appln_id
int64 7.41k
576M
| appln_filing_date
stringlengths 10
10
| docdb_family_id
int64 3.49M
82.2M
| granted
stringclasses 2
values | appln_abstract
stringlengths 120
10k
| appln_abstract_lg
stringclasses 1
value | appln_title
stringlengths 4
567
| applt_coun
stringlengths 4
289
| invt_coun
stringlengths 8
347
| cpc
stringlengths 11
1.62k
| ipc
sequence | __index_level_0__
int64 25
167k
|
|---|---|---|---|---|---|---|---|---|---|---|---|
424,011,238 | 2013-08-26 | 49,192,840 | N | A touch type naked eye three-dimensional grating and display device. The touch type naked eye three-dimensional grating comprises a touch-control electrode structure (06) located between an upper substrate (01) and a plate-shaped electrode (05). The touch-control electrode structure (06) comprises a plurality of first touch-control sensing wires (061), a plurality of second touch-control sensing wires (062) and touch-control electrodes (063). The first touch-control sensing wires (061) and the second touch-control sensing wires (062) are arranged in an intersecting mode and insulated with each other. The touch-control electrodes (063) are located in the areas defined by two adjacent first touch-control sensing wires (061) and two adjacent second sensing wires (062). Each touch-control electrode (063) and the plate-shaped electrode (05) form a capacitor body，and a touch-control function is achieved by detecting the variety of the signals on the first touch-control sensing wires (061) and the second touch-control sensing wires (062) connected with the touch-control electrodes (063). The display device comprises the above touch type naked eye three-dimensional grating. The touch type naked eye three-dimensional grating simplifies the module structure and the manufacturing process, and reduces the manufacturing cost of the whole module. | en | TOUCH TYPE NAKED EYE THREE-DIMENSIONAL GRATING AND DISPLAY DEVICE | 17940323_CN,10631781_CN | 48726851_CN,48665652_CN,16680025_CN,24097108_CN | G02B 30/25,G02B 30/27,G02B 30/28,G02B 30/31,G06F 3/0412,G06F 3/0443,G06F 3/0446,G06F2203/04103 | [
"G02F 1/13",
"G02B 30/25",
"G06F 3/041"
] | 90,256 |
50,291,755 | 1982-04-27 | 26,778,282 | Y | Apparatus for printing alpha-numeric characters for plotting an incoming signal at rates on the order of one page every seven seconds for rapidly verifying that the system producing the incoming signal is operating properly. Both printing and plotting are done by styli mounted on a moving carriage and individually electrified to produce marks on a specially prepared paper. The stylii are moved back and forth across the paper, marking in both directions. The carriage is not stepped across the page, but instead is moved smoothly from one side to the other side under control of a velocity servo. A controller generates and applies velocity command inputs to the servo, based on the position of the carriage as sensed by an optical encoder, and based on certain characteristics of the incoming signal. A first buffer memory accumulates the incoming signal until a line has been stored for plotting, while a second buffer memory is applying its accumulated contents to the printhead. Thereafter, the first memory supplies the printhead while the second memory accumulates. The apparatus includes a versatile formatting system which determines the amount of data to be plotted in each line in accordance with the incoming timing signals associated with the data. If the timing signals are not present, the apparatus makes the determination autonomously. | en | Validator for electrocardial data processing system | 5377255_US | 6039584_US | A61B 5/333,G01D 9/08,G01D 9/285 | [
"G01D 9/28",
"G01D 9/08",
"G06F 17/00"
] | 39,835 |
515,122,576 | 2018-12-07 | 64,026,470 | N | The embodiment of the invention provides a supervised word vector training method and device, and the method comprises the steps: generating the attention coefficient of each semantic vector of a target word according to the upper and lower words of the target word; generating a guide vector of the target word according to the semantic vector and the attention coefficient of each semantic vector;constructing a distance loss function from a training result of the word vector model to the guide vector; and supervising the word vector model through a distance loss function to generate an iteration direction of a training result. In the embodiment of the invention, the guide vector obtained according to the semantic vector and the attention coefficient can reflect the real semantics of the target word and the influence of the context on the semantics at the same time. Therefore, the guide vector is used for replacing the training process of the word vector model to supervise the word vector model by the upper and lower words. The obtained word vector can express the real semantics of the word, and the problems that the word vector obtained by the word vector training method in the prior art is insufficient in the aspects of expressing the polarity and the category of the word and cannot express the real semantics of the word are solved. | en | Supervised word vector training method and device | 71941820_ | 66062896_,61569412_,58883957_ | G06F 40/289 | [
"G06F 17/27"
] | 129,086 |
539,583,110 | 2019-05-27 | 67,951,389 | N | The present invention relates to a system for automatically generating text-based sentences on the basis of deep learning to achieve an improvement related to the infinity of utterance patterns and, more specifically, to a system for automatically generating, on the basis of deep learning, text-based sentences having utterance patterns of natural language. The system is configured to: receive, through a deep learning modeling unit (100), an input of utterance patterns of texts and information about common purposes of the texts and perform training for deep learning therefrom; provide, through the deep learning modeling unit, an automatic test sentence generation unit (300) with information about trained deep learning models classified according to purposes; acquire, through the automatic test sentence generation unit (300), the information about deep learning models classified according to purposes, provided by the deep learning modeling unit; input a purpose type provided through an output purpose input unit (200), to the corresponding information about deep learning models classified according to purposes and output a text-based sentence having an utterance pattern of natural language, the sentence being automatically generated according to the input purpose type; and output a credibility value for the output text-based sentence. | en | SYSTEM FOR AUTOMATICALLY GENERATING TEXT-BASED SENTENCES ON BASIS OF DEEP LEARNING TO ACHIEVE IMPROVEMENT RELATED TO INFINITY OF UTTERANCE PATTERNS | 64424550_KR | 67916813_KR,78501688_KR,68762319_KR,64320466_KR,65247966_KR,74038456_KR,67564582_KR | G06N 3/08,G10L 13/08,G10L 15/04,G10L 15/063 | [
"G10L 15/06",
"G10L 15/04",
"G10L 13/08",
"G06N 3/08"
] | 146,034 |
531,483,302 | 2019-12-23 | 70,707,438 | N | The invention belongs to the field of remote sensing image processing, and discloses a hyperspectral image dimension reduction method and device based on self-adaptive collaborative graph discriminantanalysis. The method comprises the steps of selecting part of pixels from original hyperspectral data to serve as training samples; establishing a Tikhonov regular weight coefficient matrix partitioned by categories, and constructing a collaborative representation graph; and through generalized eigenvalue decomposition, obtaining an optimal projection matrix P under an optimization criterion, andprojecting a test sample to a low-dimensional space to realize dimension reduction of the hyperspectral data. According to the method, distance-weighted Tikhonov regularization is coupled with a representation method based on l2-norm minimization, data is projected to a low-dimensional popular space, and collaborative representation characteristics are obtained through l2-norm. In the process ofconstructing the graph, the internal relation between intra-class pixels is fully mined, and cooperative representation is adaptively adjusted through distance weighting measurement. In addition, thegraph weight matrix adopts a block diagonal structure design, so that the calculation cost is reduced, and the discrimination capability is further improved. | en | Hyperspectral image dimension reduction method and device based on self-adaptive collaborative graph discriminant analysis | 58899979_ | 59214200_,70434715_,76578000_,66321312_,75244063_,64964116_ | G06K 9/6232,G06K 9/6256,G06K 9/6269 | [
"G06K 9/62"
] | 140,425 |
532,365,137 | 2019-03-14 | 70,914,325 | Y | Provided are a method for correcting cognitive bias using cognitive stimulation and a device thereof. In a method for correcting cognitive bias using cognitive stimulation performed by a computer, the method comprises the steps of : for each of a plurality of emotions subjected to correction in which emotions felt by a user are classified according to predetermined classification criteria, collecting the user′s emotion recognition information; based on the user′s emotion recognition information, evaluating the user′s emotion recognition ability for each of the plurality of emotions subjected to correction; determining the learning frequency of the user′s cognitive bias correction according to the user′s emotional recognition ability; reflecting the learning frequency for each of the plurality of emotions subjected to correction to provide context information so as to apply cognitive stimulation to the user for a predetermined period of time and receiving the learning result of the user′s cognitive bias correction; and based on the learning result of the user′s cognitive bias correction, evaluating the level of the user′s cognitive bias correction for each of the plurality of emotions subjected to correction. Accordingly, the user can correct cognitive bias for specific emotions and accurately determine the degree of cognitive bias. | en | METHOD FOR MODIFYING COGNITIVE BIAS CORRECTION USING COGNITIVE STIMULATION AND APPARATUS THEREOF | 64006636_KR | 59118083_ | A61B 5/16,A61B 5/4088,G09B 7/08,G16H 20/70 | [
"G16H 20/70",
"A61B 5/00",
"A61B 5/16",
"G09B 7/08"
] | 141,039 |
48,448,424 | 2005-07-13 | 29,709,413 | N | Clinical response to antidepressant compounds correlates with a selective down-regulation of presynaptic 5-HT1A receptors in serotonergic raphe neurons. Thus regulation of the 5-HT1A receptor gene could play a crucial role in the treatment or etiology of major depression. The promoter and repressor activities of the human 5-HT1A receptor gene have been examined. The analysis of the 5'-flanking regions of the 5-HT1A receptor gene has revealed a segment located between about -3438 and about -393 bp upstream from the initiator ATG that mediates cell-specific repression of the gene that is greater in cells that do not express the 5-HT1A receptor. The sequence of part of this region in patients with major depression were examined and a polymorphic C-G change located at -1017 bp was identified, which is associated with major depression. Thus, this sequence can be used as a genetic marker for major depression and related mental illnesses. A protein that binds to the DNA at the -1017 locus has been identified. Any such proteins that bind to the DNA at this region are important targets for the development of therapeutic compounds for the treatment of major depression and related mental illness that involve the serotonin system. In addition the promoter region from about -393 to the initiator ATG displays glucocorticoid-mediated repression. | en | Mutations of the 5' region of the human 5-HT1A gene, associated proteins of the 5' region and a diagnostic test for major depression and related mental illnesses | 5490307_CA | 6689667_CA,6689668_CA | C07K 14/72,C12Q 1/6883,G01N 33/6896,G01N2800/304 | [
"C07K 14/72",
"C07H 21/02",
"C12P 21/06"
] | 37,223 |
503,786,650 | 2017-07-18 | 59,420,192 | N | A digital piano having fixed solfège syllable keys, enabling non-stepped key changes and keypress-based sound variation, and facilitating sight singing. The piano enables continuous key variation, employs non-stepped key changes, and enables, during play, fine modulation of the pitch by means of key pressing, thus adjusting the temperament and timber accordingly. The piano has a keyboard with fixed solfège syllables; namely, each set of keys of the keyboard directly corresponds to an octave of the numbered musical notation, and each key continues to correspond to the same solfège syllable after the key is changed. A color vector image-based display based on the numbered musical notation is provided to enable a musical score to be displayed in colored images, pitches are expressed with expanded solfège syllables, colors are used to express spatial positions of the keys, notation squares having a fixed width are used to express note values, and font boldness of notation symbols is used to indicate note playing strength. The present invention provides multi-dimensional playing information, is suitable for computer-based digital score recording, has multiple display modes, and helps beginners to quickly learn to play piano by using the color vector image-based display of the numbered musical notation in combination with sight singing. | en | DIGITAL PIANO HAVING FIXED SOLFÈGE SYLLABLE KEYS, ENABLING NON-STEPPED KEY CHANGES AND KEYPRESS-BASED SOUND VARIATION, AND FACILITATING SIGHT SINGING | 64229175_CN | 63690049_CN,68333058_CN,67829148_CN,69047782_CN | G09B 15/023,G09B 15/04,G10G 1/02,G10H 1/0008,G10H 1/0016,G10H 1/34,G10H 1/344,G10H 1/44,G10H2210/395,G10H2210/471,G10H2210/541,G10H2220/096,G10H2220/116,G10H2220/221 | [
"G10H 1/00",
"G10H 1/34"
] | 122,967 |
511,233,033 | 2018-12-07 | 64,147,144 | N | The embodiment of the invention provides an unregistered word processing method based on single semantics, an intelligent question answering method and an intelligent question answering device. The unlogged-in word processing method comprises the steps that at least one semantic meaning of an unlogged-in word is obtained, and each semantic meaning comprises at least one meaning; selecting a reference semantic from the semantics of the unlogged-in words; generating semantic vectors of the unregistered words according to the reference semantics; and performing randomization processing on the semantic vector according to the random word vector to generate a simulation word vector of the unregistered word. Therefore, the simulation word vector generated through the technical scheme provided bythe invention can express the semantics of the unlogged-in words; as the randomization processing is carried out, the random variability is realized; Therefore, when the random word vector generatedby the embodiment of the invention is used for an intelligent question and answer system to generate the response, the correlation degree of the response and the question can be high, and the responsecan be diversified, so that repeated and monotonous response is avoided, the user experience is improved, and the problem of unregistered words is solved. | en | Unregistered word processing method based on single semantics, intelligent question answering method and intelligent question answering device | 71941820_ | 58883957_,61569412_,66062896_ | G06F 40/279,G06F 40/30 | [
"G06F 16/332",
"G06F 17/27"
] | 126,859 |
553,615,605 | 2019-05-25 | 66,502,274 | Y | The present invention provides a generalized consistency-based fatigue classification method for constructing a brain function network and a relevant vector machine. Compared with the prior art, the present invention has higher reliability and accuracy, and constructs an effective fatigue classification network by means of an information flowing direction and a causal relationship, so as to classify connection characteristics of a brain network in different mental states, effectively verifying results of topological structure research, and improving the ability to detect driving fatigue. The present invention constructs the brain network by means of a generalized consistency algorithm-based method, and regards the brain as a multi-regional cooperative network, so as to research the information flowing direction and the causal relationship between nodes of the brain, and to analyze changes in the topology structure of the brain network in different mental states, revealing the mechanism of fatigue generation, and providing a new perspective for fatigue-related research. The present invention uses the relevant vector machine to classify the connection characteristics, is able to achieve classification precision of over 90%, verifies the reliability of topology structure analysis, and also provides a new method for fatigue detection. | en | Fatigue Classiffication Method Based on Brain Function Network Constructed via Generalized Consistency and Relevant Vector Machine | 19913885_CN | 46664286_,80807382_,80809095_,80423800_,80041150_,80837443_,12064705_ | A61B 5/18,A61B 5/369,G06K 9/00 | [
"A61B 5/369"
] | 155,211 |
569,233,420 | 2021-12-14 | 80,755,648 | N | The invention discloses an anomaly detection method based on a heterogeneous information network element learning architecture. The anomaly detection method comprises the following steps: sampling sub-graphs from an original graph; sampling a plurality of support graph instances in the support set; iterating the sampling graph, generating structure node embedding in each sub-graph, and inputting the structure node embedding into a heterogeneous graph neural network module for training and updating; task embedding is obtained according to node embedding; calculating a support loss function and updating parameters; sampling a plurality of query graph instances, and learning node embedding and querying a loss function according to the same training process as the support set; task weights are calculated, anomaly detection is carried out, and anomaly labeling is carried out on corresponding products or nodes of the original graph. According to the invention, migration can be carried out among different heterogeneous information networks; a structure module, a heterogeneous module and a comparison module are adopted to capture structure information, heterogeneous features and unmarked information of sub-graphs respectively, and the technology is obviously superior to the latest technology on multiple heterogeneous information networks. | en | Abnormity detection method based on heterogeneous information network element learning architecture | 83763652_ | 75119326_,61172566_,59035151_,60707245_,60358103_,59844945_,60420312_,69443552_ | G06K 9/6256,G06K 9/6273,G06N 3/0445,G06N 3/0454,G06N 3/08 | [
"G06K 9/62",
"G06N 3/08",
"G06N 3/04"
] | 164,402 |
523,803,681 | 2018-12-17 | 64,101,250 | N | The present invention relates to a system for providing customized self-management information, capable of: automatically providing customized self-management information according to a user state even without performing a separate user operation or work, thereby increasing use convenience and, simultaneously, remarkably alleviating a user's emotional reactions such as depression, panic disorder, anxiety, obsession, and bipolar disorder that are caused by various types of stress; storing the user's diary, memos, and SNS content in a database, and then analyzing statistical data therefor, so as to generate and update self-paradigm information such as user patterns, habits, and tastes, thereby providing same according to a user request, such that the user can receive various pieces of information related to the user without performing a separate operation or work; and allowing a self-management application to collect texts generated by the user and images captured by the user for a preset period (T) and, simultaneously, allowing a management server to analyze the collected texts and images by using a machine learning algorithm, so as to detect user state information, and thus the present invention can increase service reliability and accuracy by automatically and accurately detecting the user state without a separate user operation. | en | SYSTEM FOR PROVIDING CUSTOMIZED SELF-MANAGEMENT INFORMATION ACCORDING TO USER STATE | 74096062_KR | 74096062_KR | G06Q 50/10,G16H 10/60,G16H 20/00 | [
"G16H 20/00",
"G16H 10/60",
"G06Q 50/10"
] | 135,154 |
17,174,766 | 1996-03-20 | 13,953,568 | Y | An artificial visual apparatus and method for image recognition having a simple adaptive scaling mechanism enables the definition of scale invariant visual icons in a processing area corresponding to the anterior inferotemporal cortex (AIT) in a one-step, value-based decision making process. Icon related activity states resulting from sensory filtering to a fourth stage KL filter (50) corresponding to the V4 area are recognized independent of the scale and position of the item to be recognized within the maximum visual field. The AIT processing area controls the window of attention in the V4 area and confines further processing onto this selected spotlight. The invention presents a biologically plausible method for scale invariant mapping from the V4 stage filter to the AIT processor. Filtering based on principal component analysis (PCA), or Karhunen-Loeve (KL) filtering, yields image data of the item of interest in the V4 stage filter, such data then being supplied to the AIT processor (70) by a scale-invariant mapping process which controls the number of inputs to the KL filters to achieve constant resolution independent of the scale of the item of interest in the maximum visual field. Thus, the problem of scale-invariant mapping is reduced to a simple adaptive thresholding by feedforward inhibition at the AIT processor. <IMAGE> | en | Scale invariance for face recognition using Karhunem-Loeve transform | 62_JP | 3587763_JP,3587762_JP,3587761_JP | G06V 40/164 | [
"G06F 15/18",
"G06N 3/00",
"G06K 9/00",
"G06T 7/00"
] | 19,194 |
450,927,707 | 2014-03-31 | 47,997,284 | N | The present invention relates to compounds of general formula I wherein R1 is halogen, lower alkyl, lower alkoxy, lower alkyl substituted by halogen, lower alkoxy substituted by halogen or cyano; R2 is hydrogen, CF3 or lower alkyl; R3 is hydrogen, lower alkyl, lower alkenyl, lower alkinyl, heterocycloalkyl, lower alkyl substitueted by cyano, cyano, benzyl substituted by halogen, 2-oxa-6-aza-spiro[3.3]hept-6-yl or is lower alkoxy substituted by halogen; X is -CH2- or -CH2-CH2-; or to a pharmaceutically acceptable acid addition salt, to a racemic mixture or to its corresponding enantiomer and/or optical isomers thereof. The compounds may be used for the treatment of schizophrenia, obsessive-compulsive personality disorder, major depression, bipolar disorders, anxiety disorders, normal aging, epilepsy, retinal degeneration, traumatic brain injury, spinal cord injury, post-traumatic stress disorder, panic disorder, Parkinson's disease, dementia, Alzheimer's disease, mild cognitive impairment, chemotherapy-induced cognitive dysfunction, Down syndrome, autism spectrum disorders, hearing loss, tinnitus, spinocerebellar ataxia, amyotrophic lateral sclerosis, multiple sclerosis, Huntington's disease, stroke, radiation therapy, chronic stress, abuse of neuro-active drugs, such as alcohol, opiates, methamphetamine, phencyclidine and cocaine. | en | PIPERAZINO[1,2-A]INDOL-1-ONES AND [1,4]DIAZEPINO[1,2-A]INDOL-1-ON E. | 7641906_CH | 13784759_,15725711_DE,44709369_ | A61K 31/4985,A61K 31/551,A61P 9/00,A61P 21/00,A61P 25/00,A61P 25/08,A61P 25/14,A61P 25/16,A61P 25/18,A61P 25/22,A61P 25/24,A61P 25/28,A61P 25/30,A61P 25/32,A61P 25/36,A61P 27/02,A61P 27/16,A61P 43/00,C07D 487/04,C07D 519/00 | [
"A61K 31/4985",
"A61P 25/00",
"A61K 31/551",
"C07D 487/04"
] | 100,179 |
4,811,549 | 2002-01-03 | 25,143,232 | N | In a combined holistic and analytic recognition system, the holistic recognition module will recognize an input word or phrase image by matching an input string of character features for the whole word or phrase against a string of prototype features for a plurality of reference words in a lexicon. This will yield a holistic answer list of recognized word or phrase candidates for the input word or phrase along with a confidence value for each answer on the list. At the same time based on each answer in the answer list, the holistic recognition modules will generate a list of character features and segment the character features into sets of each character in an answer. The analytical recognition module uses segmentation hypotheses from the segmented character feature sets to cut the image of the input string of characters into individual character images. A plurality of character images for the various segmentation hypotheses will be recognized to produce an analytical answer list having a plurality of word or phrase answers for the input word or phrase. Each analytic word answer will have a confidence value based on the combined confidence of recognizing each character. The holistic answer list and the analytic answer list will be examined to find the best answer from the two lists as therecognition of the input handwritten text. | en | HOLISTIC-ANALYTICAL RECOGNITION OF HANDWRITTEN TEXT | 13092437_US | 13092440_US,13092439_US,13092438_US | G06K 9/6292,G06V 30/1918,G06V 30/226 | [
"G06F 40/00",
"G06V 30/224"
] | 7,975 |
538,630,190 | 2020-10-01 | 68,158,900 | N | The present invention relates to sedation assessment. In order to facilitate sedation depth monitoring in an autonomous imaging setting, it is proposed to use the imaging modality itself to measure the response to suitable reflexes in order to determine the depth of sedation wherein suitable reflexes include, but are not limited to, the pupil reflex, so- called superficial reflexes and the withdrawal reflexes. In one embodiment, the pupil reflex may be measured in an MRI system by repeated interleaving of dedicated iris MR imaging with the conventional scan protocol. In another embodiment, superficial reflexes in response to stroking of the skin may be measured. This may involve a dedicated actuator that may be closely integrated with the imaging modality, e.g.an MR receive coil applied to the patient. Alternatively, remote haptic systems may be used. The reflex is then acquired with a suitable diagnostic imaging method. In another embodiment, the withdrawal reflex in response to pain may be measured. This may involve an actuator that induces sudden stitching pain or very local temperature-induced pain and that is closely integrated with the imaging modality, e.g. a pinching device integrated with a patient support or an MR receive coil applied to the patient. The reflex is then acquired with a suitable diagnostic imaging method. | en | IMAGING-BASED REFLEX MEASUREMENTS FOR SEDATION DEPTH MONITORING | 18087805_NL | 12584624_NL | A61B 3/112,A61B 3/145,A61B 5/055,A61B 5/1106,A61B 5/1128,A61B 5/163,A61B 5/4029,A61B 5/4821,A61B 6/032,A61B 6/501,A61B 6/506,A61B 9/00,G08B 6/00 | [
"A61B 5/055",
"A61B 5/00",
"A61B 5/11",
"A61B 5/16",
"A61B 6/03"
] | 145,363 |
555,667,741 | 2021-07-08 | 73,946,697 | N | The invention discloses a gesture recognition method fusing an electromyographic signal and a visual image, which comprises the following steps: acquiring an electromyographic signal by using a high-density electromyographic signal acquisition arm band, acquiring a visual image by using a camera, and fusing an electromyographic image converted from the electromyographic signal and the visual image into one image; and inputting an image fused by the surface myoelectricity image and the visual image into an image classification module for identification and classification. According to the method, the surface electromyogram signals and the visual images do not need to be identified and classified by using two different network structures, only the surface electromyogram signals and the visual images need to be transmitted into a network architecture at the same time, identification and classification can be carried out by using the same network, the network complexity can be effectively reduced, and limited resources are saved. According to the method, the form that gesture recognition is carried out by using a single input mode in each set of traditional networks can be improved, and the respective limitations that the surface electromyogram signals and the visual images are respectively used for gesture recognition are improved. | en | Gesture recognition method fusing electromyographic signals and visual image | 70489360_,75579094_ | 63708072_,61540236_,60624231_,70574549_,60692843_,60616563_,70160465_ | G06K 9/00503,G06K 9/00523,G06K 9/00536,G06K 9/6256,G06K 9/6267,G06K 9/6288,G06N 3/0454,G06N 3/084,G06V 40/113,G06V 40/28 | [
"G06N 3/08",
"G06K 9/62",
"G06N 3/04",
"G06K 9/00"
] | 156,544 |
49,663,873 | 1994-10-27 | 10,693,654 | Y | To facilitate the monitoring of spatial and haptic movement and to provide tactile feedback, the invention replaces sheath-like glove human/computer interface devices with a manually operable exo-skeletal human/computer interface device (10) which comprises a wrist encircling means (11A, 11B) and a palm encircling means (16C) and a torsionally and laterally flexible manner (16D) connecting said wrist and palm encircling means, with said wrist encircling means being adapted in use to be proximate to and generally relatively fixed around the associated wrist of one hand of a user, with said palm encircling means being adapted in use to be proximate to and relatively fixed around the palm of said hand, with said palm encircling means having an overhand member (16B) overlying the back of said hand, further comprising at least one elongate generally resilient rib member (10A, 10B) extending at a fixed end thereof from said overhand member so as to be generally coextensive with at least one finger of said hand in order to function in the form of an exoskelton thereto, adjacent a free end of said rib member at least one annular ring formation (14) which is adapted in use to receive and engage at least one finger of the said hand, and at least one position sensor of the computer system fixed relative to at least one of the said members. | en | Exo-skeletal haptic computer human/computer interface device | 7590679_US | 7185109_GB | G06F 3/014,G09B 21/003,G09B 21/009 | [
"G09B 9/00",
"G06F 3/01",
"G06F 3/00",
"G09B 21/00"
] | 39,149 |
498,674,064 | 2017-08-15 | 58,957,379 | N | Disclosed is a story flash card machine, comprising a story flash card cylinder (1), wherein the story flash card cylinder (1) comprises a rotatably arranged story flash card carrier (11), story flash cards (12) are arranged around the circumference of the story flash card carrier (11) and can rotate with the story flash card carrier (11), and the story flash cards (12) are fitted onto the story flash card carrier (11) in such a way that the story flash cards can be swung around the circumference of the story flash card carrier (11). The story flash card cylinder (1) is provided with a retention mechanism (13) to keep one of the story flash cards (12) in a specific rotating position in a specific turned-over position, so as to keep same spaced apart from an adjacent story flash card (12), so that a user can read the content in the two opened adjacent story flash cards (12). The display of the story flash cards (12) of the flash card machine in a page-turning manner inspire the brain potential of infants, improves the vision, hearing and memory of infants, promotes the development of the visual and auditory nervous systems and the brain of infants, and enhances infants' visual awareness of the contents of the story, while also improving infants' comprehensive abilities in terms of memory, observation, reaction, concentration etc. | en | STORY FLASH CARD MACHINE | 64114520_CN | 68547086_CN | G09B 5/065 | [
"G09B 5/06"
] | 120,174 |
418,200,127 | 2013-06-26 | 47,855,095 | N | A method for extended depth-of-field three-dimensional nanoscale-resolution imaging, which is applicable to the technical field of microscopic imaging, comprises the following steps: building an optical component having a double-helix point spread function and defocusing grating multistage imaging properties; performing imaging of a molecule to be detected by the optical component to obtain a double-helix image; determining a horizontal position of the molecule to be detected according to positions of middle points of double-helix sidelobes in the double-helix image; and determining an axial position of the molecule to be detected according to a rotation angle of a connection line of centers of the double-helix sidelobes in the double-helix image, and the positions of the middle points of the double-helix sidelobes. By combining the double effects of the double-helix point spread function and the defocusing grating multistage imaging, the method extends the depth of field and improves the resolution ratio. The method can be used for dynamic range imaging of subcellular fractions having any depth values in a whole cell, can produce dynamic function images of multiple moving molecules, and is of great significance to high-level understanding of the rule and relationship between subcellular structures and changes in cell function. | en | METHOD, OPTICAL COMPONENT AND SYSTEM FOR EXTENDED DEPTH-OF-FIELD THREE-DIMENSIONAL NANOSCALE-RESOLUTION IMAGING | 19477954_CN | 12550803_CN,19460246_CN,43714654_CN,47647675_CN | G01N 21/47,G01N 21/6447,G01N 21/6458,G01N 21/6486,G01N 21/84,G02B 5/1842,G02B 5/189 | [
"G01N 21/64"
] | 86,546 |
441,690,272 | 2015-01-30 | 53,406,404 | Y | The invention discloses a fluorescent diffusion optical cross-sectional image reestablishing method based on a decoupling fluorescence Monte Carlo (dfMC) model, and belongs to the technical field of biomedical engineering. The method includes the steps of firstly, determining a detecting area, selecting a plurality of scanning points in the detecting area, and obtaining fluorescent intensity distribution on a detector; secondly, establishing a three-dimensional digital model for depicting a tissue optical parameter space structure, conducting forward-direction white Monte Carlo simulation of stimulating photons according to the scanning positions and directions of a light source, tracking the stimulating photons, and recording the corresponding physical quantities of the photons reaching the detector on a path; thirdly, calculating the weight of fluorescent photons through a dfMC method, and calculating a fluorescent Jacobi matrix; fourthly, calculating the positions and absorbing coefficients of fluorophores in tissue through iterative reconstruction of GPU clusters. The method has the advantage of providing an accurate and rapid reestablishing method for a three-dimensional fluorescence tomography system through the high-precision dfMC model on the basis of the accelerated iterative reconstruction process of the GPU clusters. | en | Fluorescent diffusion optical cross-sectional image reestablishing method based on dfMC model | 11826858_ | 20113895_,39375998_,42454304_ | A61B 5/0071,G06T 17/00 | [
"G06T 17/00",
"A61B 5/00"
] | 94,221 |
477,643,569 | 2016-10-17 | 56,352,967 | Y | Provided is an omni-channel marketing curation system based on big data including a data collecting unit configured to collect behavior log data of a customer in an online store and behavior log data of the customer who visits an offline store; a big data storing unit configured to process, in real time, the log data collected in the data collecting unit; a log verbalization converting unit configured to analyze the customer's online/offline behavior log data, convert the analyzed result into one of a plurality of pre-defined verbal expressions, which indicate degrees of customer's reactions to a product, to store the converted result in the big data storing unit; a first attribute setting unit configured to receive customer's attribute information from a system user so as to create a profile for a persona according to the attribute; a persona generating unit configured to group, into one persona, customers corresponding to the attribute information received by the first attribute setting unit, generate one or more pieces of visualization information in correspondence to verbal expressions, and generate trajectory information through which how the visualization information moves to purchase the product; and a data visualization providing unit configured to consecutively play visualization information and trajectory information. | en | Omni-channel marketing curation system based on big data | 52657162_KR | 11772816_KR,52657162_KR | G06F 40/284,G06Q 30/0201,G06T 11/001,G06T 13/80 | [
"G06T 11/00",
"G06T 13/00",
"G06Q 30/02",
"G06Q 30/00",
"G06F 40/284",
"G06F 40/00",
"G06T 13/80"
] | 108,345 |
4,429,578 | 1993-01-29 | 25,525,016 | N | NITROGEN BRIDGE TETRAHYDROISOOUINOLINES Muscle relaxant nitrogen bridge tetrahydroisoquinolines are disclosed. The novel compounds are represented by the formula wherein A is -?-O- or -O-?-; M represents -(CH2)n-Z-(CH2)n or , R is a C1-3 alkoxy group, or adjacent Rs are a methylenedioxy group, R1 is lower alkyl; n is 1-6; m is 2 or 3; p is 1 - 3; Z is -N+(R2R3)-, -N(R4)-, -N(?-Y)- and -N¢(CH2)n-A-R5!-; R2 and R3 are independently lower alkyl groups wherein a carbon atom within the chain may be replaced by a heteroatom, lower cycloalkyl, lower cycloalkyl lower alkyl, aryl, aryl lower alkyl, a 4- to 6-member heteroring or may be combined with the nitrogen to form a heteroring; R4 is a straight- or branched- chain C1-10 alkyl wherein a carbon atom within the chain may be replaced by a heteroatom, lower cycloalkyl, lower cycloalkyl lower alkyl, aryl, or a heteroring, which groups may be substituted or unsubitutued; or -(CH2)n-O-?-R1; R5 is lower alkyl or lower alkenyl; Y is hydrogen, lower alkyl wherein a carbon atom within the chain may be replaced by a heteroatom, lower alkoxy, aryl, aryloxy, lower cycloalkyl, lower cycloalkyl lower alkyl, a 4- to 6-member heteroring or -NR2R3; X? is a pharmaceutically acceptable anion, and optically active forms thereof, meso forms thereof, cis-trans isomeric forms thereof and racemates thereof. | en | NITROGEN BRIDGE TETRAHYDROISOQUINOLINES | 13256734_US | 16408160_US | A61P 21/02,C07D 217/20 | [
"C07D 217/20",
"C07D 401/14",
"C07D 217/16",
"A61K 31/472",
"A61K 31/47",
"A61P 21/02"
] | 4,478 |
48,044,080 | 1998-08-11 | 26,734,369 | Y | A deck of playing cards has imprinted thereon different symbols, such as 7's, Bars, Bells, Grapes, Oranges, Melons and Cherries, and optionally a Wild Joker symbol. Each player is playing against a pay table and not against the dealer. Each player makes a first wager and the dealer deals three cards to each player. If the player is dealt three matching symbols, or One or Two Cherries or a Wild symbol, the player achieves a winning combination and is eligible to receive a payout based on the amount of the player's wager. The player may collect his winning amount and end the hand or the player may decline the winning amount, surrender his wager, make a second wager and proceed to the draw step of the method of play. If the player is dealt any pair or three-of-a-kind symbols, the player has the option of proceeding to the draw step or simply standing with his initial first wager winner or loser that was dealt to him. If the player chooses to proceed to the draw step, the player surrenders his first wager and makes a second wager for the draw step. If the player proceeds to the draw step, the player is dealt a fourth card. The player does not have to discard at any time. A player then uses the best three or four cards after the draw step to make any possible winning combination. The amount won by the player is based on a pay table. | en | Game of chance | 6365726_US,6365725_US | 6365727_US,6365728_US | A63F 1/00,A63F 3/00157,G07F 17/3293 | [
"A63F 3/00",
"A63F 1/00"
] | 36,581 |
498,237,057 | 2017-02-20 | 63,059,083 | Y | An intention understanding device based on behavior-recognition connection learning includes: an input part detecting object information of user neighborhood and joint information of user behavior, which are observed for each frame; a preprocessing part preprocessing the object information and joint information received from the input part, to enable the information to be processed with an artificial neutral network; a behavior recognition processing part classifying behavior information of the user based on the object information and joint information, which are outputted from the preprocessing part; an object relation information processing part outputting an object candidate group related with the user behavior by using the behavior information outputted from the behavior recognition processing part and the object information outputted from the preprocessing part; and an intention output part outputting a user intention recognition result through the artificial neural network inputting the behavior information outputted from the behavior recognition processing part and the object candidate group outputted from the object relation information processing part. As such, the present invention is capable of accurately predicting an intention of the user from the user behavior and the object information related with the behavior. | en | - ARTIFICIAL AGENTS AND METHOD FOR HUMAN INTENTION UNDERSTANDING BASED ON PERCEPTION-ACTION CONNECTED LEARNING RECORDING MEDIUM FOR PERFORMING THE METHOD | 68814871_KR | 64117024_KR,69157876_KR | G06F 3/011,G06N 3/084,G06V 40/16 | [
"G06K 9/00",
"G06F 3/01",
"G06N 3/08"
] | 119,907 |
51,022,558 | 2003-06-04 | 32,073,066 | N | A method and system to decode neural activity in the motor cortex to infer at least the position and velocity of a subject's hand from neural spiking activity of some number of nerve cells. In one embodiment the method includes simultaneously recording electrical activity of the nerve cells in the primary motor cortex to obtain neural data; and modeling the encoding and decoding of the neural data using a Kalman filter, where a measurement model assumes a cell firing rate to be a stochastic linear function of at least the position and velocity of the hand, and where the measurement model is learned from training data in conjunction with a system model that encodes a manner in which the hand moves. In another embodiment the method includes using the neural data to generate training data of neural firing activity conditioned on hand kinematics; learning a non-parametric representation of the firing activity using a Bayesian model; inferring an a posterior probability distribution over hand motion, conditioned on the neural training data using Bayesian inference; defining a non-Gaussian likelihood term that is combined with a prior probability for the kinematics based on learned firing models of multiple nerve cells; and using a particle filtering method is to represent, update, and propagate the posterior distribution over time. | en | Method and system for inferring hand motion from multi-cell recordings in the motor cortex using a kalman filter or a bayesian model | 6675161_US | 39069427_US,39069428_US,33378914_US,17982184_US | G06F 3/015 | [
"G06F 3/00",
"G06F 3/01"
] | 41,121 |
543,996,499 | 2019-09-26 | 69,975,485 | N | Disclosed in the embodiments of the present invention are an eye key point labeling method and apparatus, and a training method and apparatus for an eye key point detection model. The method comprises: obtaining a face image and eyelid labeling curves corresponding to the face image; for each eyelid labeling curve, integrating the eyelid labeling curve on the basis of a mathematical integral principle to determine a first curve length of an upper eyelid labeling curve and a second curve length of a lower eyelid labeling curve in the eyelid labeling curve; determining a plurality of eyelid points to be used from the upper eyelid labeling curve and the lower eyelid labeling curve; determining, on the basis of the first curve length of the upper eyelid labeling curve, the second curve length of the lower eyelid labeling curve, said eyelid points and a preset number of equally-divided points, equally-divided upper eyelid points and equally-divided lower eyelid points from the upper eyelid labeling curve and the lower eyelid labeling curve respectively, and then determining the eye key points corresponding to each face image on the basis of said upper and lower eyelid points in combination with labeled eye corner points. Therefore, eye key points having obvious semantic features are labeled, and the labeling efficiency is improved. | en | EYE KEY POINT LABELING METHOD AND APPARATUS, AND TRAINING METHOD AND APPARATUS FOR EYE KEY POINT DETECTION MODEL | 74289718_CN | 76124554_CN,78922480_CN,63499320_CN | G06V 40/193,G06V 40/197 | [
"G06T 7/00"
] | 148,433 |
442,996,324 | 2014-12-16 | 53,681,149 | N | Provided are a cognitive function evaluation apparatus, which can be used with healthy people and people with severe movement disorders, a cognitive function evaluation method, a cognitive function evaluation system, and a program. The apparatus according to the present invention is for evaluating cognitive function and is provided with a processing unit, which: analyzes brain waves arising from multiple stimulus events comprising targets and non-targets; estimates the number of stimulus events selected as targets from among the multiple stimulus events and determines the discrimination score or decoding accuracy; and evaluates cognitive function on the basis of the values of one or more of the discrimination score, the decoding accuracy, the time course of either the discrimination score or the decoding accuracy, and decoding speed. It is also possible to determine discrimination score or decoding accuracy for each different degree of difficulty for two or more stimulus event groups that differ in the degree of difficulty in discriminating between stimulus events and to evaluate cognitive function on the basis of the relationship between one or more of the discrimination score, decoding accuracy, the time course of either the discrimination score or the decoding accuracy, and the decoding speed, and said degree of difficulty. | en | COGNITIVE FUNCTION EVALUATION APPARATUS, METHOD, SYSTEM, AND PROGRAM | 64084408_JP | 64825903_JP | A61B 5/162,A61B 5/168,A61B 5/291,A61B 5/316,A61B 5/378,A61B 5/38,A61B 5/6803,A61B 5/7264,A61F 4/00,G16H 50/20 | [
"A61B 10/00",
"A61B 5/0484"
] | 94,862 |
565,817,038 | 2021-12-21 | 79,931,895 | N | The invention discloses a text emotion reason identification method based on a CPC-ANN, and belongs to the technical field of natural language processing text emotion analysis. A text emotion reason identification method based on a CPC-ANN is characterized in that a CPC-ANN model is mainly divided into five layers: an embedded layer, a Bi-LSTM layer, a Transform layer, an attention layer and a CNN output layer. A collaborative attention neural network model based on context and position interaction is provided on the basis of a CANN method, relative position information of clauses is encoded into each word vector of the clauses, and the relative position information of the clauses is encoded into each word vector of the clauses; the semantic information of the clauses is captured through a two-way long and short term memory Transformer network so as to help to generate better clause representation; then, an attention mechanism is used to capture a semantic relationship between candidate reason clauses and emotion clauses fused with context clause information; and finally, the candidate reason clauses are classified by using a convolutional neural network. According to the method, the problem that the semantic and position information is insufficiently utilized by the current text emotion reason identification method is solved. | en | Text emotion reason identification method based on CPC-ANN | 61372179_ | 60845906_,61372179_,72992234_ | G06F 40/211,G06F 40/284,G06F 40/30,G06N 3/0445,G06N 3/0454,G06N 3/084 | [
"G06N 3/08",
"G06N 3/04",
"G06F 40/211",
"G06F 40/30",
"G06F 40/284"
] | 162,681 |
49,324,657 | 2002-11-04 | 26,964,857 | N | Inclusion body myositis (IBM), the most common age-related muscle disease in the elderly population, is an incurable disorder leading to severe disability. Sporadic IBM has an unknown etiology, although affected muscles fibers are characterized by many of the pathobiochemical alterations traditionally associated with neurodegenerative brain disorders such as Alzheimer's disease (AD). Accumulation of the amyloid-beta peptide (Abeta), which is derived from proteolysis of the larger amyloid-beta precursor protein (betaAPP), appears to be an early pathological event in AD and also in IBM, where in the latter, it occurs predominantly intracellularly within affected myofibers. To elucidate the possible role of betaAPP mismetabolism in the pathogenesis of IBM, transgenic mice were derived in which betaAPP overexpression was selectively targeted to skeletal muscle using the muscle creatine kinase promoter. Skeletal muscle from transgenic mice older than 10 months was shown to contain intracellular immunoreactivity to betaAPP and its proteolytic derivatives, which was quantifiable by ELISA. In this transgenic model, selective overexpression of betaAPP leads to the development of a subset of other histopathological and clinical features characteristic of IBM, including centric nuclei, inflammation, and deficiencies in motor performance. | en | Transgenic mouse model of inclusion body myositis | 38459846_ | 38459847_US | A01K 67/0275,A01K2217/05,A01K2227/105,A01K2267/0312,A01K2267/0368,C07K 14/4711,C12N 15/8509,C12N2830/008 | [
"C07K 14/47",
"A01K 67/027",
"C12N 15/85"
] | 38,575 |
554,449,057 | 2021-05-17 | 74,922,417 | Y | The invention provides a hip joint exoskeleton control method. The hip joint exoskeleton control method comprises the following steps of acquiring a left side current rotation angle track and a left side current rotation angular velocity track of a left hip joint exoskeleton worn by a wearer, and a right side current rotation angle track and a right side current rotation angular velocity track of a right hip joint exoskeleton worn by the wearer; determining a state switching trend of the wearer according to the left side current rotation angle track, the left side current rotation angular velocity track, the right side current rotation angle track and the right side current rotation angular velocity track, wherein the state switching trend comprises stepping and standing; and according to the state switching trend, determining power-assisted torque tracks corresponding to the left hip joint exoskeleton and the right hip joint exoskeleton, and according to the power-assisted torque tracks, driving the left hip joint exoskeleton and the right hip joint exoskeleton to assist the wearer in corresponding state switching. According to the method, various different lower limb movement tasks can be accurately identified, corresponding lower limb assistance can be provided, and the method is suitable for being used in different scenes. | en | Hip joint exoskeleton control method and device | 62425052_ | 61054636_,64030769_,61116911_,58339538_,61393425_,60303373_,65424072_,58762435_,65424325_,73603239_ | A61H 3/00,A61H2003/005,A61H2003/007,A61H2201/1207,A61H2201/1642,A61H2201/165,A61H2201/5012,A61H2201/5043,A61H2201/5058,A61H2201/5069,A61H2201/5079,A61H2205/10,A61H2230/625 | [
"A61H 3/00"
] | 155,726 |
127,887 | 2002-11-22 | 8,181,317 | Y | The invention concerns the use of an H1- and/or H2-histamine receptor antagonist for the reduction of intracranial pressure (ICP), in particular for the prevention and treatment of elevated intracranial pressure and/or secondary ischaemia, in particular caused by brain injury, more in particular caused by traumatic (TBI) and non-traumatic brain injury. In particular, the histamine antagonist is selected from the group of acrivastine, alimemazine, antazoline, astemizole, azatadine, azelastine, brompheniramine, buclizine, carbinoxamine, carebastine, cetirizine, chlorcyclizine, chlorpheniramine, cinnarizine, clemastine, clemizole, clocinizine, clonidine, cyclizine, cyproheptadine, descarboethoxyloratidine, dexchlorpheniramine, dimenhydrinate, dimethindene, dimethothiazine, diphenhydramine, diphenylpyraline, doxylamine, ebastine, efletirizine, epinastine, fexofenadine, hydroxyzine, ketotifen, levocabastine, loratidine, meclizine, mequitazine, methdilazine, mianserin, mizolastine, niaprazine, noberastine, norastemizole, oxatomide, oxomemazine, phenbenzamine, pheniramine, picumast, promethazine, pyrilamine, temelastine, terfenadine, trimeprazine, tripelennamine, triprolidine, ranitidine, cimetidine, famotidine, nizatidine, tiotidine, zolantidine, ritanserine, and derivatives thereof and mixtures of any two or more of the foregoing. | en | The use of histamine receptor antagonists for the reduction of intracranial pressure | 14785_BE | 127657_ES,127659_ES,127660_ES,127655_BE,127656_ES,127653_BE,127654_BE,127652_DE,127658_ES,127661_BE | A61K 31/513,A61P 1/04,A61P 1/14,A61P 3/00,A61P 9/00,A61P 9/10,A61P 9/12,A61P 11/02,A61P 11/06,A61P 17/04,A61P 29/00,A61P 37/00,A61P 37/02,A61P 37/06,A61P 37/08,A61P 43/00,C07D 471/00,C07D 487/00,C07D 519/00 | [
"C07D 519/00",
"A61K 31/55",
"C07D 487/04",
"C07D 471/04",
"A61P 1/04",
"C07D 471/00",
"A61K 31/425",
"A61P 9/10",
"A61K 31/415",
"A61P 29/00",
"A61P 37/08",
"A61P 37/06",
"A61P 37/02",
"C07D 487/00",
"A61K 31/513",
"A61K 31/517",
"A61K 31/519"
] | 651 |
363,621,853 | 2011-10-27 | 46,050,592 | N | Provided is a sound pressure evaluation system that objectively measures UCL in a short time and/or with high precision. The sound pressure evaluation system comprises: a biosignal measuring unit that measures EEG signals of a user; a sound database that holds multiple data of pure sound; a presented sound determining unit that determines a presented sound; an output unit that presents the sound determined by the presented sound determining unit to the user as a first sound, and presents a sound with at least the same frequency and sound pressure as the first sound to the user as a second sound at a predetermined time after presenting the first sound; a volume change extracting unit that extracts the volume change between an N1 component with respect to the first sound that adopts the time at which the first sound was presented as a starting point, and that is a negative component of an event-related potential of the EEG, and an N1 component with respect to the second sound that adopts the time at which the second sound was presented as a starting point, and that is a negative component of an event-related potential of the EEG; and a validation unit that determines whether or not the sound pressure of the presented sound is too loud for the user on the basis of the volume change extracted by the volume change extracting unit. | en | SOUND PRESSURE EVALUATION SYSTEM, AND METHOD AND PROGRAM THEREFOR | 6156118_JP,32387539_,16095891_,16041597_,32381412_,43647199_ | 32381412_,32387539_,16041597_,43647199_,16095891_ | A61B 5/125,A61B 5/38,A61B 5/6814,H04R 29/001 | [
"A61B 5/0476",
"A61B 5/12",
"A61B 10/00",
"H04R 25/00"
] | 73,711 |
47,740,354 | 1998-07-15 | 26,732,584 | Y | A family of low-complexity, high quality CELP speech coders are described which use two new techniques: Backward and Inverse Filtered Target (BIFT) for fixed codebook excitation search; and Tree-Structured Multitap adaptive codebook search. Incorporation of these new techniques resulted in very low complexity CELP coders at less than 16 Kb/s. The three coefficients for linear combination of the adaptive codebook are chosen from a tree-structured tap codebook. The best tap index in the primary codebook points to a secondary codebook where the search is further conducted. This procedure may be repeated many times, wherein each subsequent tap codebook points to yet another subsequent tap codebook, which points to yet another subsequent tap codebook, etc. A fixed ternary excitation codebook using a new technique called Backward and Inverse Filtered Target matching (BIFT), is used to encode the portion of the target signal that is left behind after the adaptive codebook contribution has been subtracted. BIFT combines the elements of the Backward Filtered Target response and Inverse Filtered Target response by element-by-element multiplication to define a new vector. A predetermined number of maximums of the new vector are chosen as the pulse locations and the signs assigned are the same as the signs of the corresponding elements. | en | Low-complexity speech coding with backward and inverse filtered target matching and a tree structured multitap adaptive codebook search | 5230452_US | 6242193_US,6242192_US | G10L 19/08 | [
"G10L 19/08"
] | 36,332 |
336,625,971 | 2011-01-18 | 44,506,435 | N | Disclosed is a brain wave interface system for achieving both improvement of identification accuracy and shortening of intent relay time in item selection for a brain wave interface. A brain wave interface system is provided with a provision unit for providing a plurality of options which are operation items related to operation of an apparatus to a user; a selection blinking control unit for blinking each of the plurality of options; a biometric signal detection unit for measuring brain wave signals of the user; an estimation unit for estimating one option corresponding to a desired operation of a user using an event-related potential of brain waves included in a brain wave signal, where the event-related potential originates from blinking of each of the plurality of options; a confirmation blinking control unit for performing confirmation blinking of the one option that has been estimated; an assessment unit for assessing whether or not the estimated option is an option corresponding to the desired operation of the user, using the event-related potential of brain waves included in the brain wave signal, where the event-related potential originates from the blinking confirmation of the one selected option; and an output unit for executing processing corresponding to the assessment result of the confirmation assessment unit. | en | BRAIN WAVE INTERFACE SYSTEM, BRAIN WAVE INTERFACE PROVISION DEVICE, EXECUTION METHOD OF BRAIN WAVE INTERFACE, AND PROGRAM | 6156118_JP,32394433_,16095891_ | 16095891_,32394433_ | A61B 5/378,A61B2218/00,A61F 4/00,G06F 3/015,G06N 5/04 | [
"A61B 5/0476",
"G06F 3/01"
] | 69,327 |
52,790,967 | 1996-08-01 | 16,716,811 | Y | An electronic musical apparatus employs a tone-generator section containing a plurality of tone-generator channels, each consisting of a vowel generation unit and a consonant generation unit which operate based on formant sound synthesis, so that a song is automatically swung based on lyric data and performance data. If a syllable within words of a lyric designated by the lyric data consists of a consonant and a vowel, the consonant generation unit generates the consonant with respect to a consonant sounding time which is set in advance whilst the vowel generation unit generates the vowel to follow the consonant. If generation of multiple syllables is allocated to a desired single note within notes corresponding to a melody designated by the performance data, the multiple syllables are sequentially generated during a sounding time of the desired single note. A human operator is capable of inputting words of a lyric to form the lyric data by using a computer keyboard. Or, the human operator is capable of editing the contents of the lyric data by using the computer keyboard. Incidentally, programs and/or the lyric data and performance data can be stored in an internal memory of the electronic musical apparatus; or they can be transferred to the electronic musical apparatus from an external storage device or an external system. | en | Electronic musical apparatus using vocalized sounds to sing a song automatically | 5218280_JP | 6251162_JP,9519990_JP | G10H 1/02,G10H 1/36,G10H2220/011,G10H2250/455,G10H2250/481 | [
"G10L 13/00",
"G10K 15/04",
"G10H 1/36",
"G10H 7/00",
"G10L 13/06",
"G10H 1/02"
] | 44,698 |
45,621,305 | 2001-06-13 | 24,426,561 | N | Method and system for recognizing and characterizing bioelectric potential or electromyographic (EMG) signals associated with at least one of a coarse gesture and a fine gesture that is performed by a person, and use of the bioelectric potentials to enter data and/or commands into an electrical and/or mechanical instrument. As a gesture is performed, bioelectric signals that accompany the gesture are subjected to statistical averaging, within selected time intervals. Hidden Markov model analysis is applied to identify hidden, gesture-related states that are present. A metric is used to compare signals produced by a volitional gesture (not yet identified) with corresponding signals associated with each of a set of reference gestures, and the reference gesture that is 'closest' to the volitional gesture is identified. Signals representing the volitional gesture are analyzed and compared with a database of reference gestures to determine if the volitional gesture is likely to be one of the reference gestures. Electronic and/or mechanical commands needed to carry out the gesture may be implemented at an interface to control an instrument. Applications include control of an aircraft, entry of data from a keyboard or other data entry device, and entry of data and commands in extreme environments that interfere with accurate entry. | en | CHARACTERIZATION OF BIOELECTRIC POTENTIALS | 16597734_US | 37034199_,37034198_ | A61B 5/389,A61B 5/7264,G06F 3/015,G06V 40/20 | [
"A61B 5/0488",
"G06F 3/00",
"G06K 9/00",
"G06F 17/00",
"G06F 3/01"
] | 30,181 |
50,950,437 | 1989-12-05 | 18,026,845 | Y | A pattern recognition apparatus detects a coincidence between an input image pattern and a target pattern. Input image data fij are received, where fij denotes a concentration of the input image pattern at a picture element i,j on a two-dimensional plane of the input image pattern. Mask data guv is stored, where guv denotes a concentration of the target pattern at a point u,v on a two-dimensional plane of the target pattern. The mask data guv is scanned with respect to the input image data fij and a proportional coefficient is computed corresponding to each picture element i,j of the two two-dimensional plane of the input image pattern. The proportional coefficient denotes a ratio between the concentration of the input image data fij and the concentration of the mask data guv. The input image data fij is divided into a plurality of partial templates denoting a two-dimensional subsection of the input image pattern. For each partial template, a proportional evaluation index sigma ij is calculated corresponding to each picture element i,j of the two-dimensional plane of the input image pattern. The proportional evaluation index sigma ij denotes a degree of dispersion of the proportional coefficients within each partial template. Be examining the dispersion of the proportional coefficients, a pattern coincidence can be detected. | en | Pattern position recognizing apparatus | 5210917_JP | 8184521_JP,5931976_JP,5322912_JP | G06V 10/32,G06V 10/7515 | [
"G01B 11/00",
"G06K 9/64",
"G01R 31/28",
"G01B 11/24",
"H05K 13/08",
"G06T 1/00"
] | 40,976 |
85,857 | 2008-06-12 | 39,930,554 | N | The invention relates to a method for perception measurement, in particular for measuring the visual attention of an individual, wherein at least first visual coordinates of a first point of vision (37), associated to a first image of a visual field, and at least second visual coordinates of a second point of vision (38), associated to a second image of a visual field, are processed, first and second visual coordinates being detected essentially by a visual detection system. The invention is characterized in that the second image of a visual field is recorded after the first image of a visual field. In order to determine visual attention for specific surrounding areas as accurately as possible, the invention proposes that the second visual coordinates of the second point of vision (38) are analyzed with the first visual coordinates of the first point of vision (37) in a comparative device for meeting at least one first predetermined fixation criterium (25). When the first fixation criterium (25) is fulfilled, the first and second points of vision (37, 38) are allocated to a first fixation (48) that can be associated to ordered perception, and when the first fixation criterium (25) is not fulfilled, the first and second points of vision (37, 38) are marked and allocated to a first saccade associated with aleatoric perception. | en | METHOD FOR PERCEPTION MEASUREMENT | 12549890_AT,12549889_AT | 12549890_AT,12549889_AT | A61B 3/113,A61B 5/16,A61B 5/163 | [
"H04H 60/33",
"A61B 5/16",
"A61B 3/113"
] | 410 |
467,721,594 | 2014-09-04 | 50,049,208 | Y | Disclosed are a method and a device for expanding data of a bilingual corpus. The method for expanding data of a bilingual corpus includes: searching, in a source language-pivot language corpus, for at least one first pivot language phrase semantically matching a first source language phrase; searching, in the source language-pivot language corpus, for at least one second source language phrase semantically matching each of the first pivot language phrases to form a source language phrase set by the second source language phrases; searching, in a pivot language-target language corpus, for at least one first target language phrase semantically matching each of the first pivot language phrases to form a target language phrase set by the first target language phrases; combining the second source language phrases in the source language phrase set with the first target language phrases in the target language phrase set, so as to form at least one phrase pair in which a source language phrase and a target language phrase semantically match; and storing the formed at least one phrase pair in which the source language phrase and the target language phrase semantically match into a source language-target language corpus. Data in a bilingual corpus is expanded, so that the problem of data sparseness in the bilingual corpus is solved. | en | Method and device for expanding data of bilingual corpus, and storage medium | 45228180_CN | 53511702_CN,6729857_CN,5606466_CN,53503621_CN | G06F 16/24522,G06F 16/24556,G06F 16/3329,G06F 16/3337,G06F 40/242,G06F 40/45,G06F 40/49 | [
"G06F 17/28",
"G06F 17/30",
"G06F 17/27",
"G10L 21/00",
"G06F 40/00",
"G06F 17/21"
] | 103,011 |
274,181,600 | 2009-06-08 | 41,506,107 | Y | Systems and methods are disclosed for actively all-to-all exchanging messages in a large group of individuals, comprising accumulating messages from individuals in at least one hub, processing and reorganizing accumulated messages and their groups into short representations of the messages, and selectively actively delivering short representations according to a distribution pattern found as a maximum of a goodness function. Processing and reorganizing accumulated messages into short representations, that is summarization, is performed by constraining the stream of messages to a hierarchical stream format, having a hierarchy of messages linked by a directional parent-child relationship, and iteratively replacing the parent messages in such hierarchy with a representative message chosen among all children of that message. Distribution pattern of the messages containing short representations is found by constrained maximization of a predetermined goodness function depending on the entire distribution pattern jointly for all messages and all users. The process of summarizing messages into short representations, and the process of calculating goodness function for distribution patterns are formulated in terms of statistical model of individuals' expertise and interests, which is learnt from observing past individuals' activity. | en | Systems and methods for exchanging information in a large group | 11251605_US | 11251605_US | G06Q 10/10 | [
"G06F 15/16"
] | 59,839 |
16,820,944 | 1990-09-12 | 6,389,967 | Y | A method of this type is intended to be suitable for the recognition of individual words as well as continuous speech. It is to be characterised by robust word pattern recognition in the face of incorrect syllable segmentation and variable diction, e.g. when syllables are slurred. It must also enable fast adaptation of the system to a new speaker and an essentially unlimited generation and extension of the vocabulary from written text without explicit system training through pronunciation. An approximately real-time recognition of words and word sequences should also be possible even with very extensive vocabularies. Known methods of speech recognition require a highly expensive training procedure. Furthermore, an unmanageable flow of hypotheses is generated with continuous speech and a relatively large vocabulary, and even with medium-sized vocabularies. \n <??>According to the invention, the extracted feature vectors are segmented and classified into a set of hypotheses comprising syllable-oriented word sub-units with the aid of a word sub-unit inventory. A word hypothesis set, which is subjected to syntactic analysis in order to identify the spoken phrase, is generated from the hypothesis set by means of a three-dimensional, time-dynamic comparison with diction variants from a reference pattern vocabulary. \n <IMAGE> \n | en | Method for speaker-adaptive speech recognition | 1518_DE,708487_NL | 2898404_DE,2898403_DE | G10L 15/07 | [
"G10L 15/07"
] | 17,899 |
422,578,262 | 2014-04-02 | 51,654,820 | N | A method for facilitating a betting game based on a selected non-instantaneous event, the method comprising using at least one hardware processor for: prior to the happening of a selected occurrence in the selected non-instantaneous event, associating each player of players of the betting game with a prediction with respect to the selected occurrence, wherein the players are unfamiliar with the selected occurrence; following the commencing of the selected non-instantaneous event, receiving one or more betting antes from the players and depositing the one or more betting antes in a virtual pot; in response to receiving a betting ante of the one or more betting antes from a betting player of the players, requesting each player of the players other than said betting player to deposit an ante matching the betting ante in the virtual pot; after the happening of the selected occurrence, and for each player of the players, determining if the player is a potential winner based on a comparison between the prediction associated with the player and corresponding datum of data with respect to the selected occurrence; and if at least one player of the players is determined as a potential winner, determining at least one potential winner as a winner, and rewarding the winner with at least some of the antes accumulated in the virtual pot. | en | INTERACTIVE BETTING GAME | 48368258_GB | 48369152_IL | G07F 17/323,G07F 17/3288 | [
"G07F 17/32"
] | 89,322 |
53,157,329 | 2007-04-30 | 34,385,236 | Y | Compounds of formula (I) as defined herein: are useful for treating behavioral disorders associated with dementia, psychoses, in particular schizophrenia (deficient form and productive form) and acute or chronic extrapyramidal symptoms induced by neuroleptics; for the treatment of various forms of anxiety, panic attacks, phobias, and compulsive obsessive disorders; for treating various forms of depression, including psychotic depression; for treating disorders caused by alcohol abuse or weaning from alcohol, sexual behavior disorders, eating disorders and for treating migraine. Moreover, the compounds of the invention may be used for treating painful muscle contracture in rheumatology and in acute spinal pathology; for treating spastic contractures of medullary or cerebral origin; for the symptomatic treatment of acute and subacute pain of light to moderate intensity; for treating intense and/or chronic pain, neurogenic pain and intractable pain; for treating Parkinson's disease and Parkinson-like symptoms of neurodegenerative origin or induced by neuroleptics; for treating partial primary and secondary generalized epilepsy of simple or complex symptomology, mixed forms and other epileptic syndromes in addition to another antiepileptic treatment, or in monotherapy, for the treatment of sleep apnea, and for neuroprotection. | en | Derivatives of N-[phenyl(pyrrolidine-2-YL)methyl]benzamide and N-[(azepan-2-YL) phenylmethyl]benzamide, preparation method thereof and application of same in therapeutics | 5223465_FR,9766164_FR,7973352_FR,5834715_FR,5686367_FR | 9766164_FR,7973352_FR,5686367_FR,5834715_FR | A61P 3/04,A61P 11/00,A61P 15/00,A61P 15/08,A61P 19/00,A61P 25/00,A61P 25/02,A61P 25/04,A61P 25/06,A61P 25/08,A61P 25/16,A61P 25/18,A61P 25/22,A61P 25/24,A61P 25/28,A61P 25/32,A61P 29/00,C07D 207/09,C07D 223/04 | [
"C07D 233/04",
"A61K 31/445"
] | 45,455 |
559,326,084 | 2020-07-27 | 70,656,748 | N | Provided are a voice activity detection method and apparatus for emotion recognition, an electronic device, and a storage medium, relating to artificial intelligence. The method comprises: acquiring an audio signal (S1); performing a processing operation on the audio signal, the operation comprising: adding a pure noise segment and a human voice noise segment under multiple scenes to the audio signal and randomly setting a signal-to-noise ratio (S2); extracting MFCC characteristics of the processed audio signal and second-order difference characteristics thereof (S3); inputting the characteristics into a neural network model, and extracting high-dimensional information of the audio signal and adjacent correlation characteristics of the audio signal (S4); inputting the extracted high-dimensional information and correlation characteristics of the audio signal into a fully connected network model to obtain a detection result of each frame of the audio signal, the detection result comprising human voice and non-human voice (S5); and segmenting, according to the detection result of the audio signal, the audio signal into a human voice part and a non-human voice part (S6). The voice activity detection method can accurately implement voice activity detection in a low signal-to-noise ratio environment and a non-stable environment. | en | VOICE ACTIVITY DETECTION METHOD AND APPARATUS FOR EMOTION RECOGNITION, ELECTRONIC DEVICE, AND STORAGE MEDIUM | 64024668_CN | 71870857_CN,79089723_CN | G10L 25/03,G10L 25/24,G10L 25/30,G10L 25/63,G10L 25/84,G10L 25/87 | [
"G10L 25/84",
"G10L 25/63",
"G10L 25/24",
"G10L 25/03",
"G10L 25/87",
"G10L 25/30"
] | 159,069 |
557,865,597 | 2020-04-03 | 71,188,858 | N | The present disclosure relates to a multi-target electrical stimulation circuit, an electrical stimulator, and a signal output method therefor. The electrical stimulation circuit comprises a control module, multiple electroencephalography collecting modules, and multiple stimulation adjusting modules. Each electrode is correspondingly provided with one electroencephalography collecting module and one stimulation adjusting module. Different electrodes are used for stimulating different targets. The electroencephalography collecting modules are used for collecting electroencephalography signals on the corresponding electrodes and transmitting to the control module. The control module is used for acquiring a brain rhythm phase signal on the basis of an electroencephalography signal received, and outputting a stimulation signal at a set waveform phase point when the brain rhythm phase signal is phase-locked. The stimulation adjusting module is used for adjusting, on the basis of the stimulation signal received, an electroencephalography stimulation signal outputted to the corresponding electrodes. The technical solution of the present invention favors the implementation of electrical stimulation with respect to multiple targets, and implements time-locked matching of electrical stimulation with respect to the multiple targets. | en | MULTI-TARGET ELECTRICAL STIMULATION CIRCUIT, ELECTRICAL STIMULATOR, AND SIGNAL OUTPUT METHOD THEREFOR | 78687075_CN | 82942661_CN,64523646_CN | A61B 5/369,A61B 5/4094,A61B 5/7257,A61B2576/026,A61N 1/36064,A61N 1/36139 | [
"A61N 1/36",
"A61B 5/00"
] | 158,057 |
556,352,213 | 2021-05-14 | 77,415,710 | N | The invention discloses a group relationship quality evaluation and intervention method and system based on a graph pairing task. The method comprises the following steps: establishing a test task of a two-person cooperation puzzle; the method comprises the following steps: carrying out double-person cooperation task testing on pairwise testees in a group in turn, recording test data and electroencephalogram signals, calculating to obtain task performance indexes and related characteristics of the electroencephalogram signals when the pairwise testees cooperate to complete tasks, and carrying out classified prediction evaluation on double-person relationship quality and contradictions by combining a machine learning classification model. Obtaining quality scores and conflict levels of pairwise tested relations in the group; pairwise subjects with high contradiction level in the group are extracted for adaptive feedback training, and the contradiction level is reduced to a set range; and averaging all the relation quality scores obtained by the pairwise subjects to obtain an evaluation score of the group relation quality. According to the method, objective and accurate evaluation of the group relationship is realized, and the purpose of closed-loop intervention is achieved in combination with the group relationship quality. | en | Group relationship quality evaluation and intervention method and system based on graph pairing task | 59003648_ | 61053047_,67470834_,60730174_ | A61B 5/7264,G06K 9/00523,G06K 9/6247,G06Q 10/06393 | [
"G06K 9/62",
"G06Q 10/06",
"G06K 9/00",
"A61B 5/372"
] | 157,157 |
484,077,343 | 2017-04-05 | 59,870,975 | N | The invention aims to govern illegal information and illegal small advertisements, which are also known as urban psoriasis. The invention aims to govern the existing problems, and to govern and solve the existing problems fundamentally. The illegal small advertisements and people manufacturing the illegal small advertisements do not have a deep relationship, contact ways on the advertisements are generally commonly-used phone numbers, and thus the starting point of the invention is that the problem can be completely solved only by taking a principle, of who using the contact ways pay, for governance regardless of who manufacture the illegal small advertisements or who use the contact ways. The following difference points used for charges are that: the different charges are adopted according to different information and different locations on the advertisements, and the charges of companies and enterprises and individuals are also different, the costs of the companies and the enterprises need to be higher, and need to be as higher as times compared with the costs of normal individual advertising. Scrolling of daily interest bearing and monthly compound interest is carried out 12 times in a year, basic money amounts need to be settled, and day/month interest after the settlement is determined according to governance effects. | en | Governance of illegal information and illegal small advertisements | 67134549_ | 67134549_ | G06Q 30/0241,G06Q 50/26 | [
"G06Q 50/26",
"G06Q 30/02"
] | 111,947 |
510,735,643 | 2019-03-25 | 66,290,520 | N | A temporal prediction model (222) for semantic intent understanding is described. An agent (202) (e.g. a moving object such as a pedestrian, an animal, a cyclist, a vehicle, etc.) in an environment can be detected in sensor data collected from sensors on a vehicle. The sensors can include light detection and ranging (LIDAR) sensors, radio detection and ranging (RADAR) sensors, ultrasonic transducers, sound navigation and ranging (SONAR) sensors, location sensors (e.g., GPS, compass, etc.), inertial sensors (e.g., inertial measurement units, accelerometers, magnetometers, gyroscopes, etc.), cameras, wheel encoders, microphones, environment sensors. Computing device(s) associated with the vehicle determine, based partly on the sensor data, attribute(s) of the agent (e.g., classification, position, velocity, etc.), and generate, based partly on the attribute(s) and a temporal prediction model (222), semantic intent(s) of the agent (e.g., crossing a road, staying straight, etc.), which correspond to candidate trajectory(s) (238) of the agent. The candidate trajectory(s) (238) are associated with weight(s) representing likelihood(s) that the agent will perform respective intent(s). The computing device(s) uses one (or more) of the candidate trajectory(s) (238) to determine a vehicle trajectory along which a vehicle is to drive. | en | TEMPORAL PREDICTION MODEL FOR SEMANTIC INTENT UNDERSTANDING | 55150187_US | 74058768_US,74195026_US | B60W 50/0098,B60W 60/00274,B60W2050/0025,B60W2420/52,B60W2552/45,B60W2554/402,B60W2554/4026,B60W2554/4029,B60W2554/4041,B60W2554/4042,B60W2554/806,B60W2556/40,G01S 17/58,G01S 17/89,G01S 17/931,G05D 1/0088,G05D 1/0221,G05D 1/0246,G05D 1/0255,G05D 1/0257,G05D 1/0274,G05D2201/0213,G06N 5/02,G06N 20/00,G08G 1/166 | [
"G08G 1/16",
"G05D 1/00",
"G01S 17/58"
] | 126,667 |
560,716,913 | 2020-04-01 | 71,649,536 | N | The invention relates to a retina stimulator with grid vision detection. The retina stimulator comprises an image processing device and an implantation device; the image processing device carries out graying processing on the initial image to obtain a grayscale image, obtains a histogram of the grayscale image according to the grayscale image, obtains feature information of the grayscale image according to the grayscale image and the histogram, and judges whether the grayscale image is a grating vision image or not according to a first type of preset threshold value; if the gray level image is a grating vision image, binarization processing is carried out to obtain a binary image; the binary image and a plurality of preset mapping lines obtain the number of corresponding pixel points with target pixel value projections on each preset mapping line, the target preset mapping line is determined based on the number of the corresponding pixel points with the target pixel value projections and a second type of preset threshold, and the bar grating direction of the initial image is determined; a visual signal is transmitted to the implant device based on the determined grid direction; and the implant device converts the visual signal into an electrical stimulation signal to stimulate ganglion cells or bipolar cells of the retina. | en | Retina stimulator with grid vision detection function | 60112516_ | 58378007_,58323229_,71077150_ | G06T 7/77 | [
"G06T 7/77",
"G06K 9/46"
] | 159,876 |
335,853,529 | 2011-01-11 | 44,304,167 | N | A recognition task execution means (11) for providing a feature point selection system capable of selecting suitable feature points that fit a recognition algorithm for a recognition task, for the plurality of evaluation-use images, executes the recognition task while using the significance of each of a plurality of feature point candidates upon a three-dimensional shape model. A recognition error margin evaluation means (12) evaluates a recognition error margin related to all evaluation-use images based on a difference between recognition results of the recognition task and correct response data of the recognition task for each of the evaluation-use images. A feature point significance determination means (13) defines, for recognition error margins related to all evaluation-use images, a cost function represented by a function to which a constraint condition has been added for causing the significance of insignificant feature point candidates to approach zero easily, and calculates a significance for each feature point candidate that is likely to minimize the value of the cost function. A feature point selection means (14) selects feature points to be used for the recognition task from feature point candidates upon the three-dimensional shape model on the basis of the significance of each of the feature point candidates. | en | FEATURE POINT SELECTION SYSTEM, FEATURE POINT SELECTION METHOD AND FEATURE POINT SELECTION PROGRAM | 16702485_JP,5232977_JP,32452576_JP,42666828_JP,41920505_JP | 41920505_JP,32452576_JP,42666828_JP,16702485_JP | G06K 9/6255,G06T2207/20081,G06T2207/30201,G06V 10/44,G06V 20/64,G06V 40/175 | [
"G06T 1/00",
"G06T 7/00"
] | 68,733 |
563,066,662 | 2021-01-25 | 78,962,590 | N | The invention provides an eyestrain recovery device, a stereoscopic image generation method and device, and a recording medium, which can promote the recovery of eyestrain when a user views an image which enables the user to perceive the sense of far and near. An eyestrain recovery device is provided with: an eyepiece for a user to observe; a display having a display surface for displaying the sighting target; and an image generation unit that generates a left visual target image and a right visual target image. The image generation unit performing the steps of: determining positions on the display surface corresponding to the center position of the eyepiece as a left reference position and a right reference position, respectively; a display unit that displays a left background image and a right background image each having a left reference position and a right reference position as substantially centers, respectively, and that displays areas other than the left background image and the right background image with a predetermined color; solving a position PosL and a position PosR; solving a left display position and a right display position; and respectively displaying a left sighting target image and a right sighting target image for forming the sighting target at the left display position and the right display position. | en | Eye fatigue recovery device, stereoscopic image generation method and device, and recording medium | 81979242_,84093515_ | 84907228_,60299400_,82014685_,84015082_,72944752_,82501428_,82228209_,82025966_ | A61H 5/00,A61H 5/005,A61H2201/1207,A61H2201/50,A61H2201/5043 | [
"A61H 5/00"
] | 161,376 |
528,924,124 | 2019-12-30 | 67,081,185 | N | The invention discloses an image depth estimation algorithm based on deep learning and Fourier domain analysis. The image depth estimation algorithm comprises the steps: reading an image pair in a data set, and inputting the image pair into a proposed network architecture to obtain a reconstruction view; substituting the image pair and the reconstructed view into a loss function, and training andconfirming a network architecture to obtain an initial disparity map; substituting the image pair clipped according to the clipping ratio lambda into the confirmed network architecture to obtain a parallax mapping candidate object; and fusing the parallax mapping candidate objects, and estimating the depth and displaying the depth on a display device, wherein the image pair area is divided into aleft view and a right view, and the reconstructed view area is divided into a reconstructed left view and a reconstructed right view. According to the image depth estimation algorithm, based on the network architecture of deep learning and the Fourier domain analysis unsupervised monocular image estimation deep learning algorithm, the precision of the depth map and the precision of moving image depth estimation are effectively improved, and the accuracy and robustness of the algorithm are improved through the improved network architecture. | en | Image depth estimation algorithm based on deep learning and Fourier domain analysis | 59079142_ | 72971166_,58832091_ | G06T 7/55 | [
"G06T 7/55"
] | 138,787 |
544,341,016 | 2020-10-21 | 74,109,004 | N | The invention provides an RPA and AI combined data processing method and device, equipment and a storage medium, and relates to the technical field of RPA and AI. The method comprises the following steps: acquiring text data; determining speaking contents and context information corresponding to the speaking content according to the text data; on the basis of the context information correspondingto the speaking content and a trained language model, determining speaker information and emotion information corresponding to the speaking content; on the basis of the speaker information and the emotion information corresponding to each speaking content, determining a trained target acoustic feature model corresponding to each speaking content; and on the basis of each speaking content and the target acoustic feature model corresponding to each speaking content, generating sound data corresponding to the text data and outputting the sound data. Due to the fact that the identity gender and the speaking emotion of the speaker are comprehensively considered in the generation of the sound data, the generated sound data can more accurately reflect the role and emotion of the speaker of the reading material, the accuracy of the sound generation of the audio reading material is effectively improved, and the user experience is improved. | en | RPA and AI combined data processing method and device, equipment and storage medium | 59189179_,70163463_ | 73642030_,60290270_,64889497_,72823247_ | G06K 9/6223,G10L 13/033,G10L 13/047,G10L 13/08 | [
"G10L 13/047",
"G06K 9/62",
"G10L 13/08",
"G10L 13/033",
"G10L 13/04"
] | 148,963 |
48,475,495 | 2005-07-20 | 34,727,611 | N | A noninvasive measurement system for monitoring activity condition of a living body is provided, which can accurately monitor activity condition of a desired region of a living body. The means for measuring activity information is equipped with the light emitting source unit 3 to emit a light beam toward a living body, the light detecting unit 4 to receive the light beam scattered by blood present in the head 10, the temporary operation unit 1 A to compute temporary activity condition of blood in the head 10 by performing arithmetical operation of signals of the light beam to be received in the light detecting unit 4, the superficial activity operation unit to compute activity condition of blood in the scalp 14, and the correction operation unit 1 B to compute corrected activity condition of the brain 11 by correcting temporary activity condition of the head 10 that is computed in the temporary operation unit 1 A based on computed activity condition of blood in the scalp 14 in the superficial activity operation unit. Further, the means for measuring positional information is equipped with the positional information collecting unit to monitor positional information of the brain 11 in the head 10, and the memory unit 8 to memorize positional information of the brain 11 obtained in the positional information collecting unit. | en | Noninvasive measurement system for monitoring activity condition of living body | 5988941_JP | 5274900_JP | A61B 5/0261,A61B 5/055,A61B 5/14553 | [
"A61B 5/026",
"A61B 5/00",
"A61B 5/0285",
"A61B 5/055",
"A61B 5/01",
"A61B 10/00"
] | 37,271 |
510,996,384 | 2018-12-07 | 64,001,970 | N | The embodiment of the invention provides a multi-semantics-based extraset word processing method and device. The method comprises the following steps: obtaining the weight of each semantic meaning ofan extraset word according to the words of the extraset word in a sentence; generating a semantic vector of each semantic meaning according to the word vector of the prosthetic meaning in each semantic meaning; and according to the weight of each semantic meaning, carrying out weighted summation on the semantic vectors of each semantic meaning to generate a simulation word vector. According to thesimulation word vector generated by the technical scheme provided by the invention, other semantics of the extraset words can be considered while sentence meaning matching is realized, so that semantics expressed by the simulation word vector are richer and fuller, and the simulation word vector adapts to a richer semantic environment; when the simulation word vector is used in the intelligent interaction system, the correlation degree of the response and the question is high, the response accuracy is improved, the system adapts to a richer dialogue environment, the intelligent question and answer system is more intelligent, the user experience is greatly improved, and the problem of foreign word collection in the prior art is solved. | en | Multi-semantics-based extraset word processing method and device | 71941820_ | 66062896_,61569412_,58883957_ | G06F 40/284,G06F 40/30 | [
"G06F 17/27",
"G06F 16/33",
"G06F 16/332"
] | 126,796 |
4,882,171 | 2003-10-17 | 32,108,080 | N | Certain atypical antipsychotic medications (particularly clozapine) or combinations of medications are useful to treat alcohol or other substance abuse, particularly in the general (non-schizophrenic) population. Generally stated, one aspect of the invention features a method of treating a patient suffering from alcohol or other substance abuse by administering to the patient medication effective to rectify an abuse-associated dysfunction in t he DA-mediated brain reward circuit. A second aspect of the invention features administering medication that strongly antagonizes .alpha.2 andrenergic receptors and weakly antagonizes dopamine D2 receptors. Preferably, the rati o of .alpha.2 receptor blockade to D2o receptor blockade is similar to that of clozapine. The medication may be a single compound (such as clozapine or risperidone), or it may include two or more compounds which together achieve the specified function. For example, the medication may include a first component which weakly blocks the D2 receptor (such as clozapine, quetiapine or ziprasidone or a low dose of another anti-psychotic that is a more potent D2 blocker) and a5 second component (such as clozapine, risperidone or idazoxan) which strongly blocks .alpha.2 receptors, particularly the .alpha. 2C receptor. Cocktails of the twocomponents are also disclosed. | en | TREATING ALCOHOL AND OR SUBSTANCE ABUSE BY ANTAGONIZING .ALPHA. 2 ADRENERGIC RECEPTORS WITH WEAK DOPAMINE BLOCKING | 6104054_US,16834881_US | 16834883_US,16834884_US,16441702_US,16834882_US | A61K 31/415,A61K 31/5513,A61P 25/30,A61P 25/32 | [
"A61K 31/55",
"A61K",
"A61K 31/415"
] | 8,870 |
416,216,940 | 2013-02-22 | 50,188,910 | Y | Abstract An information classification program includes the following: accepting a search keyword; retrieving, from multiple information items (111) posted by multiple users, a posted information item (111) including the accepted search keyword, each of the multiple posted information items (111) including at least either of a text information item (113) and an image information item (112), and acquiring posted information items (111) which are within a predetermined chronological range with respect to the posted information item (111) including the search keyword; and classifying, as image information items (112) related to the search keyword, some of image information items (112) included in the posted information items (111) that have been acquired, and performing first determination of, for each of the classified image information items (112), whether or not a user who posted an information item (111) including the classified image information item (112) took an action related to the search keyword. CD C co-) co , co S0 0 w 0 0 ~-j 0 0 -U U ) u) u ~~~~ <: C 0HWH o U)< < U0 <0 0 0i 0 U)f <L m 0 0 -j -... - i C) C) Q < w C: CD~~C CNCOI- L o r- 0 F C\CJ w-J 0- 0 F-- _F- LLF- 0 0IU 0 D CD L zUD ~ ~ H 5U Ed 0 0D < O = a_ co Cl) 0~~ 0 0 D0 =D~ H-: - = LL 05 Of CID WZ - L 0 _L 0D L WU)CI LLU H0 H0 LD < = a ~ 0 0 W0-HU) C) 0 | en | Information classification program, information classification method, and information processing apparatus | 80712450_JP | 47682504_,43524649_,43524648_ | G06F 16/30,G06F 16/35,G06F 16/58 | [
"G06F 17/27"
] | 85,411 |
508,702,345 | 2017-12-01 | 60,551,843 | N | A head-mounted immersive virtual reality display device and an immersive virtual reality display method. The head-mounted display device comprises a head tracking and orientation detection system (1), a head-mounted optical display system (2), an image segmentation system (3), an encoding and decoding system (4), and a processing system (5). The head tracking and orientation detection system (1) is used to capture an image of a peripheral environment of a human eye. The encoding and decoding system (4) decodes an image signal of the head tracking and orientation detection system (1) before transmitting the decoded image signal to the processing system (5). Once the same has been processed by the processing system (5), the encoding and decoding system (4) performs encoding, and image information is transmitted to the image segmentation system (3). The image segmentation system (3) divides the received image information into a plurality of display images, and then the head-mounted optical display system (2) performs display. The head tracking and orientation detection system (1) captures an image of a peripheral environment and acquires head position and orientation information, and superimposes virtual video information over the image information of the peripheral environment, thereby combining the real and virtual world. | en | HEAD-MOUNTED IMMERSIVE VIRTUAL REALITY DISPLAY DEVICE AND IMMERSIVE VIRTUAL REALITY DISPLAY METHOD | 64526632_CN | 67856991_CN,71863023_CN,71641586_CN,67622487_CN | G02B 27/0101,G02B 27/017,G02B2027/0138,G02B2027/014,H04N 13/344 | [
"G02B 27/01"
] | 125,766 |
475,220,346 | 2016-07-14 | 56,555,296 | Y | SURGICAL STAPLING LOADING UNIT WITH STROKE COUNTER AND A surgical loading unit (10), which comprises: an elongate outer frame (12) defining a longitudinal axis (k) and having proximal and distal ends (14, 16); an end effector (300 mounted to the distal end (16) of the outer frame (12); a drive (22) at least partially disposed within the outer frame (12) and operatively coupled to the end effector (300), the drive (22) adapted for longitudinal movement through a plurality of sequential firing strokes to operate the end effector (300); a rotatable counter (32) mounted to the outer frame (12) and adapted for rotational movement about the longitudinal axis (k), the rotatable counter (32) having visual indicators for providing visual indicia corresponding to a number of firing strokes completed by the drive (22); a counter actuator (52) mounted to the drive (22) and dimensioned to engage the rotatable counter (32) to cause rotational movement of the rotatable counter (32) through an arc segment of rotation during each firing stroke of the drive (22); and a cam actuator (54) mounted to the drive (22) and dimensioned to operatively engage the counter actuator (52) to cause release of the counter actuator (52) from the rotatable counter (32) upon longitudinal movement of the drive (22) through each successive return stroke. C-4- | en | Surgical stapling loading unit with stroke counter and lockout | 12334718_US | 54969210_ | A61B 17/068,A61B 17/07207,A61B 17/105,A61B2017/00473,A61B2090/0803,A61B2090/0807,A61B2090/0814 | [
"A61B 17/068",
"A61B 17/072"
] | 106,794 |
457,055,783 | 2015-09-22 | 53,840,459 | N | An automatic unattended-monitoring visual acuity inspection device is disclosed. The inspection device includes an inspection panel, an inspector (2) and an isolation wall (3) arranged opposite to the inspection panel. The inspection panel includes a liquid crystal display panel (12), a lower polarizer (13) and a backlight source (14). The liquid crystal display panel (12) is configured to display visual display pictures. The isolation wall (3) is provided with a wireless signal transmitter configured to transmit a visual acuity indication signal collected from a person to be inspected to the wireless signal receiver through a network. The wireless signal receiver is configured to transmit the received visual acuity indication signal to a processor. The processor is configured to process the visual acuity indication signal and transmit a process feedback to the inspection panel. The device may prevent achieving a high level of visual acuity of the person to be inspected by cheating behaviors, such as remembering position information regarding an inspection chart or reminding the person to be inspected by other persons, thereby highly inspecting a real visual acuity level of the person to be inspected. The device is suitable for applying in those occasions where a strict requirement for visual acuity inspection is needed. | en | AUTOMATIC VISION TESTING DEVICE WITHOUT HUMAN MONITORING | 49191785_CN | 49139024_CN,54463474_CN,54451480_CN | A61B 3/0025,A61B 3/0033,A61B 3/032,A61B 3/066,G02F 1/133528,G06T 7/70,G06V 40/10,H04N 7/183 | [
"A61B 3/032",
"G02F 1/1335",
"H04N 7/18",
"G06T 7/70",
"G06K 9/00",
"A61B 3/00",
"A61B 3/06"
] | 101,713 |
543,048,870 | 2020-03-30 | 71,003,428 | Y | The present disclosure provides a method and equipment for searching a non-unique solution of a petrophysical property combination in history matching. The method includes: receiving field historical data, and dividing the field historical data into a tuning historical data portion, and a validation historical data portion; inputting initial data for the petrophysical property combination into an proxy model to obtain field response data; optimizing the petrophysical property combination to minimize an error between the field response data corresponding to the optimized petrophysical property combination and the field response data of the tuning historical data portion; inputting the optimized petrophysical property combination into the proxy model to obtain field response data corresponding to the optimized petrophysical property combination; matching the field response data corresponding to the optimized petrophysical property combination and the field response data of the validation historical data portion; and if the field response data corresponding to the optimized petrophysical property combination and the field response data of the validation historical data portion match each other, determining the optimized petrophysical property combination as the non-unique solution of the petrophysical property combination. | en | Method and equipment for searching non-unique solution of petrophysical property combination in history matching | 52486354_CN,53520583_CN,77783141_CN,54631115_CN | 53520583_CN,77783141_CN | E21B 41/00,E21B 49/00,E21B2200/20,E21B2200/22,G06F 30/27,G06F 30/28,G06F2111/10,G06N 3/006,G06N 3/126,G06N 5/003,G06N 5/02,G06N 20/00 | [
"G06N 5/00",
"G06F 111/10",
"G06N 5/02",
"G06N 20/00",
"G06F 30/27",
"E21B 49/00",
"E21B 41/00"
] | 147,975 |
445,009,581 | 2015-03-13 | 50,946,995 | N | An intravascular interventional tactile probe having a touching force range and position information feedback, comprising: a central column (7); a plurality of side wall electrodes (2) arranged on the periphery of the central column (7); a touch sleeve (1) which has electrical conductivity and can move relative to the central column (7) in an axial direction and in a radial direction; a top touch electrode (3) which can be movably connected to the central column (7); and an early-warning electrode (4) which can generate a side touch position signal when the touch sleeve (1) moves relative to the central column (7) in the radial direction and is electrically connected to at least one of the plurality of side wall electrodes (2). When moving for a first distance (d1) relative to the central column (7) in the axial direction, the touch sleeve (1) can be electrically connected to the top touch electrode (3) so as to generate a top touch signal, and when the axial displacement of the touch sleeve (1) relative to the central column (7) reaches the second distance (d2), the touch sleeve (1) can be electrically connected to the early-warning electrode (4) and generate an early-warning signal. The tactile probe having the above structure can distinguish a touch position and can indicate the magnitude of a touching force as well. | en | INTRAVASCULAR INTERVENTIONAL TACTILE PROBE HAVING TOUCHING FORCE RANGE AND POSITION INFORMATION FEEDBACK | 63504924_CN | 63628725_CN,63721587_CN | A61B 5/6885,A61B 5/7455,A61B 5/746,A61B 90/00,A61B 90/06,A61B2090/065,A61M 25/01,A61M 25/0105,A61M2025/0166,A61N 1/05 | [
"A61M 25/01",
"A61B 19/00"
] | 96,153 |
571,825,046 | 2021-05-31 | 74,834,226 | N | A machine reading comprehension method and apparatus based on BERT, and a computer device and a computer-readable storage medium, which relate to the technical field of artificial intelligence and the technical field of neural networks. The method comprises: training a first pre-set pre-training language model according to a question-document pair to be trained, so as to generate a document sorting model (S102); training a pre-set multi-document answer prediction model according to the question-document pair to be trained, so as to generate a reading comprehension model (S103); on the basis of the document sorting model and according to a question-document pair to be predicted, outputting a target document corresponding to a second question (S105); and on the basis of the reading comprehension model and according to the second question and the target document, acquiring target text in the target document that is output by the reading comprehension model, and taking the target text as a reading comprehension answer to the second question (S106). A multi-document reading comprehension question is converted into a single-document reading comprehension question, so as to reduce the interference of answer extraction during reading comprehension, thereby improving the accuracy of a multi-document reading comprehension answer. | en | MACHINE READING COMPREHENSION METHOD AND APPARATUS BASED ON BERT, AND DEVICE AND STORAGE MEDIUM | 63942312_CN | 63487878_CN,67862676_CN | G06F 40/166,G06F 40/205,G06F 40/295 | [
"G06F 40/205"
] | 165,656 |
4,780,560 | 2001-05-18 | 27,394,794 | Y | This invention relates to an apparatus and method for assessing a subject's hearing by recording steady-state auditory evoked responses. The apparatus generates a steady-state auditory evoked potential stimulus, presents the stimulus to the subject, senses potentials while simultaneously presenting t he stimulus and determines whether the sensed potentials contain responses to t he stimulus. The stimulus may include an optimum vector combined amplitude modulation and frequency modulation signal adjusted to evoke responses with increased amplitudes, an independent amplitude modulation and frequency modulation signal and a signal whose envelope is modulated by an exponential modulation signal. The apparatus is also adapted to reduce noise in the sens ed potentials by employing sample weighted averaging. The apparatus is also adapted to detect responses in the sensed potentials via the Phase weighted T- test or Phase zone technique. The apparatus may further perform a number of objective audiological tests including latency tests, AM/FM discrimination tests, rate sensitivity tests, aided hearing tests, depth sensitivity tests, supra-threshold tests and auditory threshold tests. The apparatus is further adapted to perform multi-modality testing in which more than one sensory modality of the subject is tested simultaneously. | en | SYSTEM AND METHOD FOR OBJECTIVE EVALUATION OF HEARING USING AUDITORY STEADY-STATE RESPONSES | 13054591_CA | 13054592_CA,13054593_CA | A61B 5/0051,A61B 5/121,A61B 5/38,A61B 5/7257,H04R 25/70 | [
"H04R 25/00",
"H04R 29/00"
] | 7,604 |
273,710,839 | 2008-04-21 | 39,671,992 | N | The present invention relates to the single use of the progesterone-recep tor antagonist 11.beta.-(-4-acetylpheny)-17 .beta.-hydroxy-17.alpha.-(1,1,2, 2,2-pentafluoroethyl)- estra-4,9-dien-3-one or a pharmaceutically acceptable derivative or analogue thereof for the prophylaxis and treatment of BRCA1- or BRCA2-mediated breast cancer, as well as to a combination comprising the progesterone- receptor antagonist 11 .beta.-(4-acetylphenyl)-17.beta.-hydrox y-17.alpha.-(1,1,2,2,2- pentafluoroethyl)-estra-4,9-dien-3-one or a pharmace utically acceptable derivative or analogue thereof, together with at least o ne pure antiestrogen, for the prophylaxis and treatment of BRCA1- or BRCA2-m ediated breast cancer, ovarian cancer endometrial cancer, gastric cancer, co lorectal cancer.endometriosis, myeloma, myoma and meningioma. The present in vention also relates to a combination of the progesterone-receptor antagonis t 11 .beta.-(4-acetylphenyl)-17.beta.-hydroxy-17.alpha.-(1,1,2,2,2-pentafluo roethyl)-estra-4,9- dien-3-one or a pharmaceutically acceptable derivative o r analogue thereof, together with at least one EGF or EGFR targeting drug, f or the prophylaxis and treatment of BRCA1- or BRCA2-mediated breast cancer, ovarian cancer endometrial cancer, gastric cancer, colorectal cancer, endome triosis, myeloma, myoma and meningioma. | en | PROGESTERONE-RECEPTOR ANTAGONIST FOR USE IN BRCA ALONE OR ASCOMBINATIONWITH ANTIESTROGEN | 5288655_DE | 12529801_DE,12529800_DE | A61K 31/00,A61K 31/567,A61K 31/573,A61K 45/06,A61P 5/24,A61P 5/32,A61P 5/36,A61P 15/00,A61P 35/00,A61P 43/00 | [
"A61P 5/32",
"A61K 45/06",
"A61K 31/573",
"A61P 5/36",
"A61K 31/567",
"A61K 31/00"
] | 59,208 |
568,214,774 | 2021-06-29 | 73,392,217 | N | Disclosed in the present invention is an MFCC feature encoding-based combined ultrawideband cross polarization chipless RFID tag, comprising a tag patch unit, a dielectric substrate, and a grounding layer. The tag patch unit comprises a barcode type resonant unit and a double L-shaped resonant unit. The barcode type resonant unit is formed by arranging five identical rectangular patches in parallel and rotating same counterclockwise. The double L-shaped resonant unit is formed by reversely combining two L-shaped patches that are composed of four identical rectangular patches. A transmitting antenna transmits a horizontally polarized electromagnetic wave as an interrogation signal; scattered waves of the signal reflected by the tag are obtained by a receiving antenna; a receiver obtains a spectrum of the scattered waves; the spectrum is converted into a time domain signal by means of inverse Fourier transform; a response to the tag is extracted by means of a window; an MFCC feature is extracted by means of steps such as pre-emphasis and short-time Fourier transform. A 2-8 GHz MFCC feature encoding mode is achieved by means of the tag, thereby achieving encoding of 15 bits in total. The present invention has the advantages of being low in cost, large in encoding capacity, and easy to be detected in the actual environment. | en | MFCC FEATURE ENCODING-BASED COMBINED ULTRAWIDEBAND CROSS POLARIZATION CHIPLESS RFID TAG | 64526632_CN | 78483536_CN,85283806_CN,76447368_CN,84696989_CN | G06K 19/07745,G06K 19/07773 | [
"G06K 19/077"
] | 163,944 |
491,018,317 | 2016-09-05 | 58,230,456 | N | A three-dimensional face reconstruction method based on grayscale and depth information, mainly comprising: performing identification with respect to grayscale information of a face; performing identification with respect to depth information of the face; performing multi-modal face identification; performing a matching operation by means of a 3D model; and performing 3D reconstruction of the face. The method comprises the processes of: locating a feature area with face data and performing registration and feature extraction by using a feature point; using the Adaboost algorithm to select a most effective feature for classification; using a nearest neighbor classifier to calculate a matching score so as to realize the multi-modal face identification; and completing face reconstruction by matching a local 3D model. The method effectively improves the performance and the efficiency of a face identification system by using a fused strategy. By utilizing 3D-based cascade regression, and selecting a dense 3D point set, the present invention enables a face to be completely marked to prevent a position change of a marker, thus solving the issues of movement-caused positioning point inconsistency and self-closure. The present invention reduces the computational costs, is highly versatile, and has superior real-time performance. | en | THREE-DIMENSIONAL FACE RECONSTRUCTION METHOD BASED ON GRAYSCALE AND DEPTH INFORMATION | 63936558_CN | 64524059_CN | G06T 17/00,G06T2200/04,G06V 20/653,G06V 40/165,G06V 40/172 | [
"G06T 17/00"
] | 115,912 |
530,284,013 | 2019-08-07 | 70,161,923 | Y | Abstract Certain aspects involve using a space-time memory network to locate one or more target objects in video content for segmentation or other object classification. In one example, a video editor generates a query key map and a query value map by applying a space-time memory network to features of a query frame from video content. The video editor retrieves a memory key map and a memory value map that are computed, with the space-time memory network, from a set of memory frames from the video content. The video editor computes memory weights by applying a similarity function to the memory key map and the query key map. The video editor classifies content in the query frame as depicting the target feature using a weighted summation that includes the memory weights applied to memory locations in the memory value map. 200 ,, 202 Access a queryframe Generate a query key map and a Generate memory key maps and 204'e query value map from features of the 205 memory value maps for memory query frame frames 206 Retrieve memory key maps and memory value maps from a memory network Compute memory weights by applying a similarity function to the memory key 208 maps and the query key map Generate a segmentation mask for the query frame from a weighted 210 summation that includes the memory weights applied to the memory value maps | en | NON-LOCAL MEMORY NETWORK FOR SEMI-SUPERVISED VIDEO OBJECT SEGMENTATION | 71788041_US | 37151460_,54962077_,75387897_ | G06K 9/627,G06V 10/26,G06V 10/454,G06V 10/764,G06V 10/82,G06V 20/41,G06V 20/44,G06V 20/46,G06V 20/48,G06V 20/49,H04N 21/432,H04N 21/44,H04N 21/44008,H04N 21/47205,H04N 21/47217 | [
"G06T 7/10",
"G06K 9/46"
] | 139,654 |
273,954,948 | 2009-06-15 | 41,433,621 | N | Methods and diagnostic kits are disclosed which, are useful for determining whether a subject has a likelihood or propensity to develop a neurodegenerative disease, or for diagnosing a neurodegenerative disease in a subject The method includes (a) quantitating the amount of alpha-synuclein and total protein in a cerebrospinal fluid (CSF) sample obtained from the subject and calculating a ratio of alpha-synuclein to total protein content; (b) comparing the ratio of alpha-synuclein to total protein content in the CSF sample obtained from the subject with the alpha-synuclein to total protein content ratio in CSF samples obtained from healthy subjects that at the time of sampling did not show clinical signs of neurodegenerative disease and that did not develop neurodegenerative disease; and (c) determining, from the comparison in step (b), whether the subject has a likelihood to develop neurodegenerative disease or making a diagnosis of neurodegenerative disease in a subject. A difference in the ratio of alpha-synuclein to total protein content in the CSF sample of the subject when compared with the ratio of alpha-synuclein to total protein content in the CSF samples obtained from healthy subjects indicates that the subject has a likelihood to develop a neurodegenerative disease or has developed a neurodegenerative disease. | en | METHODS AND KITS FOR DIAGNOSING NEURODEGENERATIVE DISEASE | 25700625_DE,40996780_CA,40269792_CA,41250255_AE | 40269792_CA,25700625_DE,41250255_AE | G01N 33/6896,G01N2333/47,G01N2800/2814,G01N2800/2821,G01N2800/2828,G01N2800/2835,G01N2800/387 | [
"G01N 33/68",
"G01N 33/577",
"G01N 33/543",
"G01N 33/532"
] | 59,456 |
482,304,726 | 2017-04-01 | 59,483,360 | N | The present invention provides an intelligent drawing toy for children. The intelligent drawing toy comprises a drawing interface and a base. The drawing interface is connected with the base; the drawing interface comprises three parts of the left part, the middle part and the right part; the left part of the base is provided with control elements required by the left part of the drawing interface; a drawing area is arranged in the middle part of the base; the right part of the base is provided with control elements required by the right part of the drawing interface; and the remaining base cavity is provided with a controller, a data memory, a power switch, a battery slot, and a power charging socket. The intelligent drawing toy for children provided by the present invention is more in line with the sequential characteristics of the psychological development of children, the intuitive color selection keyboard and the graphic selection keyboard can better stimulate the drawing interest of the children, and each adjustment knob has the advantages of easy operation and flexible use; and according to the intelligent drawing toy for children provided by the present invention, without using a pencil or a magnetic pen, the phenomenon of children stab wound by pen tips can be avoided, and the painted image can be easily saved. | en | Intelligent drawing toy for children | 64984313_ | 65015668_,58280966_,59401758_ | A63H 33/22,A63H 33/26,G06F 3/03545,G06F 3/0412,G06F 3/04842,G06F 3/04845,G06F 3/04883 | [
"G06F 3/041",
"G06F 3/0354",
"A63H 33/22",
"G06F 3/0484",
"A63H 33/26",
"G06F 3/0488"
] | 110,769 |
575,401,988 | 2021-12-30 | 82,218,708 | N | Materials-by-design is a new paradigm to develop novel high-performance materials. However, finding materials with superior properties is often computationally or experimentally intractable because of the astronomical number of combinations in design spaces. The disclosure is a novel AI-based approach, implemented in a game-theory based generative adversarial neural network (GAN), to bridge the gap between the physical performance and design space. A end-to-end deep learning model predicts physical fields like stress or strain directly from the material geometry and microstructure. The model reaches an astonishing accuracy not only for predicted field data but also for secondary predictions, such as average residual stress at R2˜0.96). Furthermore, the proposed approach offers extensibility by predicting complex materials behavior regardless of shapes, boundary conditions and geometrical hierarchy. The deep learning model demonstrates not only the robustness of predicting multi-physical fields, scalability, and extensibility. The disclosure may alter physical modeling and simulations by incorporating material geometry and boundary conditions into a graphical representation, and vastly improves the efficiency of evaluating physical properties of hierarchical materials directly from the geometry of its structural makeup. | en | End-to-End Deep Learning Approach to Predict Complex Stress and Strain Fields Directly from Microstructural Images | 5217896_US | 72391755_US,83574253_TW,83658990_US | B29C 64/386,B33Y 50/00,G06F 30/20,G06N 3/0454,G06N 3/088,G06T 11/00 | [
"G06N 3/04",
"G06N 3/08",
"G06T 11/00",
"B29C 64/386",
"B33Y 50/00",
"G06F 30/20"
] | 167,291 |
411,111,326 | 2012-04-19 | 46,621,430 | N | Disclosed is a double-path power feedback system for a laser processing device. Besides that a laser output power detection circuit (10) is arranged at an output end side of a laser (2), a processing point is provided with a processing point laser output power detection circuit (6), whose input end is connected with an output end of a laser emitting unit (5) through a splitting and coupling lens, and whose output end is separately connected with a negative feedback control circuit (9) having a feedback switchover function and a laser power supply control circuit, and is used for detecting laser output power of at least one processing point in real time, and outputting a negative feedback thereof to a control end of the laser power supply control circuit in real time, thereby implementing double-path power negative feedback, so as to control the laser output power. The system can significantly improve the laser output stability and the laser power control accuracy, obviously improve the laser processing quality, and effectively solve the problem of unstable output power caused by that a single power negative feedback closed-ring control system of a laser process device does not comprise a laser incidence unit (3), an optical fiber (2) and a laser emitting unit (5) of an output optical fiber coupling transmitting system. | en | DOUBLE-PATH POWER NEGATIVE FEEDBACK SYSTEM FOR LASER PROCESSING DEVICE | 12604641_CN,46800739_CN,46800738_CN,46800737_CN | 12604641_CN,46800738_CN,46800739_CN | B23K 26/03,B23K 26/06,H01S 3/13 | [
"H01S 3/131",
"H01S 5/0683",
"B23K 26/70",
"B23K 26/00"
] | 82,621 |
567,763,921 | 2021-12-22 | 77,531,452 | N | The invention discloses a children education method based on an OID technology, and belongs to the field of intelligent education, and the education method comprises the following specific steps: (1) arranging a programming robot; (2) collecting surrounding environment information; (3) confirming the advancing route and starting to move; (4) problem identification and voice broadcast are carried out; (5) receiving answers; (6) analyzing answers and broadcasting whether the answers are correct or not; (7) after the route is finished, stopping moving and broadcasting a result; in the step (1), the user moves the programming robot to a designated position and starts the programming robot by turning on an operation switch; according to the invention, the action speed of the robot can be improved, the situation of insufficient answering or missing answering of questions is avoided, the user experience is improved, the integrity of children education is ensured, the children education quality is improved, meanwhile, proper prompts are given to facilitate the cultivation of children's thinking ability, the comprehensive literacy education of children's worldwide understanding, science understanding and art enlightenment is improved, and the interestingness of children is improved. And the treatment modes are more diversified. | en | Child education method based on OID technology | 65248800_ | 65248800_ | G09B 7/02,G09B 7/04 | [
"G09B 7/04",
"G09B 7/02"
] | 163,741 |
543,978,170 | 2019-06-27 | 67,919,794 | N | A dynamic function mode learning method inspired by an fMRI brain network mechanism, comprising: collecting resting-state fMRI brain image data of several subjects; performing a preprocessing operation on the resting-state fMRI brain image data; according to the preprocessed resting-state fMRI brain image data, using a GICA-IR method to separately obtain a group-level resting-state brain function network and an individual-level resting-state brain function network, as well as time sequences corresponding thereto; using a sliding time window method to calculate a dynamic function connection matrix between resting-state brain function networks corresponding to each subject, and stretching triangular elements on the matrix into dynamic function connection vectors, so as to obtain a set of dynamic function connection vectors corresponding to all of the subjects; and using a deep neural network model and an affine propagation clustering algorithm to extract a dynamic function connection pattern intrinsic to the brain that is implied in the set of dynamic function connection vectors. The described method provides a solid basis for revealing the basic principles of cognitive activities of the brain and the damage mechanism of cranial nerves diseases, and exploring the features of occupational brain plasticity reorganization. | en | DYNAMIC FUNCTION MODE LEARNING METHOD INSPIRED BY FMRI BRAIN NETWORK MECHANISM | 69044620_CN | 78936227_CN,79087803_CN,78963968_CN,78961946_CN,66818898_CN,68598172_CN | G06K 9/6218,G06N 3/0445,G06N 3/0454,G06N 3/08,G16H 50/50 | [
"G06N 3/08",
"G06K 9/62",
"G06N 3/04",
"G16H 50/50"
] | 148,295 |
410,477,073 | 2012-03-19 | 49,161,385 | N | The present invention relates to a method for measuring a cognitive distortion of a three-dimensional image. The measurement method according to the present invention comprises: (a) a step of selecting any one from a plurality of three-dimensional parameters for the scale for cognizing a three-dimensional image in a three-dimensional manner; (b) a step of displaying the test stereoscopic image corresponding to the selected three-dimensional parameter on a screen; (c) a step of displaying, on a screen, a plurality of sample images obtained by varying the value of the selected three-dimensional parameter for the test stereoscopic image; (d) a step of selecting at least one of the plurality of sample images; and (e) a step of performing the above-described steps (b) to (d) at least two times on the different test stereoscopic images, and measuring a three-dimensional cognitive scale for the selected three-dimensional parameter based on the sample image selected in step (d). Thus, whether a viewer distortedly cognizes a three-dimensional image, or the degree to which a viewer cognizes the three-dimensional image in connection with the three-dimensional parameter such as the size, depth and the like of the three-dimensional image can be measured by means of the three-dimensional cognitive ability of each individual viewer. | en | METHOD FOR MEASURING COGNITIVE DISTORTION OF THREE-DIMENSIONAL IMAGE | 46740573_KR,12641820_KR | 46740573_KR | G06T 3/0093,G06T 7/97,G06T2200/04,G06T2207/10012,H04N 13/00,H04N 13/106,H04N 13/30,H04N 17/00,H04N 17/004,H04N2013/0074 | [
"G06T 15/00",
"H04N 13/00"
] | 82,198 |
51,683,465 | 2000-03-31 | 26,825,517 | Y | The present invention provides a method for determining whether an individual has a central nervous system disease or injury comprising: obtaining a T2 measurement for at least one brain tissue sample from an individual; determining if the T2 measurement indicates the presence of a central nervous system disease or injury in the individual. The present invention also provides a method for determining whether an individual has a central nervous system disease or injury comprising: obtaining a first T2 measurement using an MRI technique for a brain tissue sample from an individual at a first time; obtaining a second T2 measurement using an MRI technique for the brain tissue sample from the individual at a second time; and comparing the second T2 measurement to the first T2 measurement. In addition, the present invention provides a method for determining the effectiveness of a treatment for a central nervous system disease or injury comprising: obtaining a first T2 measurement using an MRI technique for a brain tissue sample from an individual at a first time; administering a first treatment to an individual; obtaining a second T2 measurement using the MRI technique for the brain tissue sample from the individual after administering the first treatment; and comparing the second T2 measurement to the first T2 measurement. | en | Method for determining activity in the central nervous system | 6633360_US | 7210284_US,8567960_US,8567957_US | A61B 5/055,A61B 5/411,A61B 5/7264 | [
"A61B 5/055"
] | 42,582 |
405,967,354 | 2012-05-14 | 48,167,093 | N | The invention provides a time-resolved single-photon or ultra-weak light multi-D imaging spectral system and a time-resolved single-photon or ultra-weak light multi-D imaging spectral method. On the one hand, in order to realize rough time resolution, the invention provides a time-resolved single-photon counting 2D imaging system for forming color or grey imaging. Moreover, in order to realize high-precision time resolution, the multi-D imaging system provided in the invention mainly comprises a light source, an imaging spectral measurement unit, an electric detection unit, a system control unit and an algorithm unit, wherein the light carrying the information of an object is imaged on a spatial light modulator and randomly modulated according to the compressed sensing theory, the emergent light of a grating is collected by using a point or array single-photon detector, meanwhile, the number of photons and the photon arrival time are recorded, and reconstruction is carried out by using the compressed sensing algorithm and the related algorithm of the spectral imaging. The system has the advantages of single-photon detection sensitivity, high time resolution and wide spectral range, and can be widely applied in the areas of numerous new high-tech industries such as unicellular biophysics and material defect detection. | en | TIME-RESOLVED SINGLE-PHOTON OR ULTRA-WEAK LIGHT MULTI-DIMENSIONAL IMAGING SPECTRUM SYSTEM AND METHOD | 49112595_CN | 49415859_CN,49811528_CN,49379839_CN,49501862_CN,49303400_CN,49130062_CN | G01J 1/0414,G01J 1/4228,G01N 21/17,G01N 21/255,G01N 21/6456,G01S 7/4816,G01S 7/486,G01S 17/89,G02B 26/0833,H03K 21/38 | [
"G01J 1/42",
"G02B 26/08",
"G01N 21/25",
"G01N 21/17",
"G01N 21/64",
"G01S 7/48",
"G01S 17/89"
] | 79,616 |
560,826,420 | 2020-07-16 | 72,116,901 | N | Provided is an image content automatic description method based on the construction of a Chinese visual vocabulary list. The method comprises steps performed in order: step a, using a Chinese word segmentation tool to perform word segmentation processing on several descriptive sentences corresponding to a single picture, selectively reserving nouns, verbs and adjectives in a word list according to statistical word frequencies, and then using the reserved words to form a Chinese visual vocabulary list; step b, carrying out prediction on the Chinese visual vocabulary list on the basis of a Chinese vocabulary list prediction network, to obtain image label information; and step c, on the basis of an automatic image description model, using an encoder to extract image convolutional features, and then using a decoder to decode the image convolutional features, as an initial input, into a Chinese descriptive statement. Image label information can be obtained by carrying out prediction on an image vocabulary list on the basis of a vocabulary list prediction network, and a residual structure is added to a Chinese visual vocabulary list prediction network, such that the problem of network degradation along with an increase in the number of layers of a Chinese visual vocabulary list prediction network can be effectively solved. | en | IMAGE CONTENT AUTOMATIC DESCRIPTION METHOD BASED ON CONSTRUCTION OF CHINESE VISUAL VOCABULARY LIST | 82734136_CN | 17960227_CN,18179363_CN,17973463_CN,20494245_CN | G06F 40/216,G06F 40/284,G06K 9/6267,G06N 3/0454,G06N 3/049,G06N 3/08,G06V 10/40 | [
"G06F 40/284"
] | 159,925 |
416,081,019 | 2013-12-06 | 50,250,166 | N | The invention relates to a high precision micro-electromechanical combined inertial navigation unit which comprises a global positioning system (GPS) receiving module, an angular velocity sensor, an acceleration sensor, low pass filters connected with the acceleration sensor, high-speed analog-digital conversion modules, a digital signal processor comprising software program, and a corresponding power supply; the high precision micro-electromechanical combined inertial navigation unit is characterized by further comprising a field-programmable gate array (FPGA); three axes of micromechanical gyroscopes are respectively connected with the corresponding high-speed analog-digital conversion modules by the low pass filters; three axes of micromechanical accelerometers are respectively connected with the corresponding high-speed analog-digital conversion modules by the low pass filters, and are respectively connected with the FPGA. The FPGA can be programmed into a plurality of communication interfaces and digital filters, and all the functions in the FPGA can be completed in parallel. The GPS receiving module is directly connected with the FPGA. The high precision micro-electromechanical combined inertial navigation unit is small in volume, light in weight, good in real-time performance, low in cost and high in accuracy. | en | High precision micro-electromechanical combined inertial navigation unit | 20123307_ | 20123307_ | G01C 21/165,G01S 19/47 | [
"G01S 19/49",
"G01C 21/18"
] | 85,299 |
530,894,825 | 2020-03-09 | 70,528,503 | N | The invention discloses a low-complexity sparse Bayesian vector estimation method and system. The method comprises the following steps: S1, setting initial values of dictionary matrix posteriori distribution and additive white Gaussian noise probability distribution hyper-parameters in a sparse Bayesian vector system; S2, on the basis of a preset system model and the initial values of the hyper-parameters, acquiring estimated values of dictionary matrix posteriori distribution and additive white Gaussian noise probability distribution hyper-parameters in the system by using an edge likelihoodmaximization algorithm; and S3, estimating a sparse vector of the system based on the estimated value of the algorithm probability distribution hyper-parameter. According to the low-complexity sparseBayesian vector estimation method and the low-complexity sparse Bayesian vector estimation system, a posterior probability model of a sparse vector is constructed and a sparse vector omega estimationproblem can be converted into an estimation problem of hyper-parameters d and sigma 2, and then a sparse Bayesian vector estimation algorithm is adopted to solve a problem that matrix inversion is needed in a common sparse Bayesian estimation algorithm, so that the reliable estimation performance is obtained with relatively low complexity. | en | Low-complexity sparse Bayesian vector estimation method and system | 64955811_ | 61071308_,60806845_,75087443_,60240137_,62363818_,65094953_ | G01S 3/14 | [
"G01S 3/14"
] | 140,102 |
445,679,529 | 2015-10-30 | 54,361,101 | N | The invention relates to a device for suppressing a pathologically synchronous and oscillatory neuron activity, comprising: a stimulation unit, which can be implanted in the body of a patient, for stimulation, by way of electrical and/or optical stimuli, of neurons in the brain and/or spinal cord of the patient, said neurons exhibiting pathological synchronous and oscillatory neuron activity and said stimuli being intended to suppress the pathological synchronous and oscillatory neuron activity when administered to the patient; a measuring unit for recording measurement signals reflecting the neuron activity of the stimulated neurons; and a control and analysis unit for controlling the stimulation unit and for analyzing the measurement signals. Said control and analysis unit is configured in such a way that it controls the stimulation unit such that the stimulation unit administers stimuli, checks the success of stimulation on the basis of the measurement signals recorded in response to the application of the stimuli, and, if the stimulation success is not sufficient, inserts one or more stimulation breaks in the application of the stimuli or extends one or more stimulation breaks, no stimuli which could suppress the pathological synchronous and oscillatory neuron activity being applied during the stimulation breaks. | en | DEVICE FOR EFFECTIVE INVASIVE DESYNCHRONIZING NEUROSTIMULATION | 7501357_DE | 52979561_GR,41625658_DE,16884276_DE | A61B 5/24,A61B 5/4836,A61N 1/025,A61N 1/0551,A61N 1/3605,A61N 1/36064,A61N 1/36067,A61N 1/36082,A61N 1/36096,A61N 1/36103,A61N 1/36125,A61N 1/36135,A61N 1/36139,A61N 1/36178,A61N 5/0622,A61N 5/067,A61N2005/0612,A61N2005/0629,A61N2005/0651 | [
"A61N 1/36",
"A61N 5/06"
] | 96,801 |
520,325,196 | 2019-03-26 | 66,286,040 | N | A method for automatically training a neural network is disclosed. The method including: at a trainer having a first communication device and a perception recorder, continuously recording the surroundings in the vicinity of the first object; receiving, at the trainer, a message from a communication device associated with an object in the vicinity of the trainer, the message including information about the position and the type of the object; identifying a recording corresponding to the time at which the message is received from the object; correlating the received positional information about the second object with a corresponding location in the recording to identify the object in the recording; classifying the identified object based on the type of information received in the message from the object; and using the classified recording to train the neural network. RECORD PERCEPTION OF SURROUNDING 202 AREA AND PROVIDE TO COMMUTING RECEIVE COMMUNICATION FROM OBJECT -204 (INCLUDING TYPE, POSITION) IDENTIFY A RECORDING CORRESPONDING TO THE TIME WHEN COMMUNICATION IS -,-206 CORRELATE RECEIVED POSITIONAL -11208 INFORMATION WITH CORRESPONDING LOCATION IN RECORDING TO IDENTIFY CLASSIFY THE IDENTIFIED OBJECT BASED 21 ON TYPE INFORMATION RECEIVED IN /210 COMMUNICATION FORWARD CLASSIFIED RECORDING TO o212 CLASSIFIER/NEURAL N/W | en | Systems and methods for automatically training neural networks | 14948445_AU | 72654996_,72764397_ | B60W 40/09,B60W 50/00,B60W2540/30,B60W2554/00,G01S 7/417,G01S 13/00,G01S 13/86,G01S 13/865,G01S 13/867,G01S 13/89,G01S 13/931,G01S 15/931,G01S 17/89,G01S 17/931,G01S2013/9316,G01S2013/9323,G01S2013/9324,G05D 1/0246,G06F 1/12,G06K 9/00536,G06K 9/6256,G06K 9/6267,G06N 3/008,G06N 3/08,G06V 10/82,G06V 20/10,G06V 20/56,G06V 20/58,G06V 40/10,G08G 1/162,G08G 1/163,H04L 67/12,H04W 4/00,H04W 4/02,H04W 4/185,H04W 4/38,H04W 4/40,H04W 4/46 | [
"G01S 13/86",
"G08G 1/00",
"G06N 3/08",
"G06N 3/067",
"G06K 9/00",
"B60W 40/09"
] | 132,703 |
4,282,657 | 1988-02-29 | 22,202,301 | Y | 64680-442 PIPERAZINYL-HETEROCYCLIC COMPOUNDS Arylpiperazinyl-ethyl(or butyl)-heterocyclic compounds and their pharmaceutically acceptable acid addition salts are neuroleptic agents. They are useful in the treatment of psychotic disorders and have the following formula: <IMG> wherein Ar is naphthyl optionally substituted by fluoro, chloro, trifluoromethyl, methoxy, cyano or nitro; quinolyl; isoquinolyl; 6-hydroxy-8-quinolyl; benzoisothiazolyl or an oxide or dioxide thereof each optionally substituted by fluoro, chloro, trifluoromethyl, methoxy, cyano, or nitro; benzothiazolyli benzothiadiazolyl; benzotriazolyl; benzoxazolyl; benzoxazolonyl; indolyl; indanyl optionally substituted by one or two fluoro; 3-indazolyl optionally substituted by 1-trifluoromethylphenyl; or phthalazinyl; n is 1 or 2; and X and Y together with the phenyl to which they are attached form quinolyl; 2-hydroxyquinolyl; 64680-442 benzothiazolyl; 2-aminobenzothiazolyl; benzoisothiazolyl; indazolyl; 3-hydroxyindazolyl; indolyl; spiro¢cyclopentane-1,3'-indolinyl!; oxindolyl; optionally substituted by one to three of (C1-C3) alkyl, or one of chloro, fluoro or phenyl, said phenyl optionally substituted by one chloro or fluoro; benzoxazolyl; 2-aminobenzoxazolyl; benzoxazolonyl; 2-aminobenzoxazolinyl; benzothiazolonyl; benzoimidazolonyl; or benzotriazolyl. | en | PIPERAZINYL-HETEROCYCLIC COMPOUNDS | 6385626_US | 12648971_US,16180907_US | A61P 25/18,A61P 25/20,C07D 209/34,C07D 235/26,C07D 263/58,C07D 277/68,C07D 401/06,C07D 413/12,C07D 417/12 | [
"C07D 295/04",
"A61P 25/20",
"A61K 31/495",
"C07D 277/68",
"C07D 263/58",
"C07D 209/34",
"A61K 31/496",
"A61P 25/18",
"C07D 209/08",
"C07D 263/56",
"C07D 209/36",
"C07D 249/18",
"C07D 215/12",
"C07D 209/40",
"C07D 215/38",
"C07D 215/40",
"C07D 235/26",
"C07D 417/00",
"C07D 409/06",
"C07D 401/06",
"C07D 403/06",
"C07D 413/12",
"C07D 413/00",
"C07D 417/06",
"C07D 417/12",
"C07D 413/06"
] | 3,768 |
4,649,837 | 1999-11-17 | 22,736,102 | N | Provided herein is a non-invasive method of treating, controlling or preventing medical, psychiatric or neurological disorders, using transcutaneous electrical stimulation. The method employs a plurality of stimulation frequency parameters, ranging from a relatively high frequency, for example about 40,000 Hertz, to a relatively low frequency, for example about 250 Hertz, the: entire plurality of frequency parameters being administered at each of a plurality of stimulation intensity levels. In particular, the method involves stimulating at a first highest frequency parameter and a first lowest intensity parameter, incrementally decreasing the: stimulation frequency parameter a lowest frequency parameter, increasing the frequency parameter to the highest frequency parameter and increasing the intensity parameter to a next highest intensity parameter, and again stimulating through the plurality of frequency parameters from the highest frequency to the lowest frequency. The method described herein is useful in treating, controlling and/or preventing various disease states and disorders, including without limitation, tremor disorders, such as essential tremor and Parkinson's disease, dementia disorders, such as Alzheimer's disease and painful degenerative disorders, such as reflex sympathetic dystrophy and fibromyalgia. | en | METHOD AND APPARATUS FOR TREATING CHRONIC PAIN SYNDROMES, TREMOR, DEMENTIA AND RELATED DISORDERS AND FOR INDUCING ELECTROANESTHESIA USING HIGH FREQUENCY, HIGH INTENSITY TRANSCUTANEOUS ELECTRICAL NERVE STIMULATION | 16616694_US | 13037323_US | A61N 1/36021,A61N 1/36025 | [
"A61N 1/36",
"A61N 1/34"
] | 6,039 |
571,825,179 | 2021-05-31 | 74,354,070 | N | The present application relates to the technical field of artificial intelligence. Disclosed are a three-dimensional point cloud semantic segmentation method and apparatus, and a device and a medium. The method comprises: performing, by using a pre-set spatial cell, point cloud division and quantitative discrimination on three-dimensional point cloud data to be subjected to prediction, so as to obtain target point cloud data; inputting the target point cloud data into a point cloud semantic category prediction model for semantic category probability prediction, so as to obtain a point cloud semantic category probability prediction value of the target point cloud data, wherein the point cloud semantic category prediction model is a model obtained by means of training on the basis of a PointSIFT neural network module and a PointNet++ neural network; and determining a target semantic category of each point in the target point cloud data according to the point cloud semantic category probability prediction value. Fast and accurate logical division for point clouds of a complex large-scale target object is realized, thereby improving the recognition precision of point cloud segmentation; and fine features of the complex target object can be better processed, thereby improving the accuracy of semantic category prediction. | en | THREE-DIMENSIONAL POINT CLOUD SEMANTIC SEGMENTATION METHOD AND APPARATUS, AND DEVICE AND MEDIUM | 63942312_CN | 82742475_CN,69638817_CN,63958634_CN | G06K 9/6267,G06N 3/0454,G06N 3/08,G06V 10/267,G06V 10/44,G06V 10/462,G06V 10/56 | [
"G06K 9/62",
"G06N 3/04"
] | 165,658 |
544,184,384 | 2019-08-23 | 74,086,604 | Y | The present invention relates to an apparatus for measuring a disorder part by using a cortical-function brain network and a method thereof. According to the present invention, The apparatus comprises: a receiver for receiving an electroencephalogram signal measured from a plurality of nodes closely contacting the scalp of a subject in a resting state; a control unit which calculates a brain network value for the degree and strength of the connection between interconnected nodes by using the received electroencephalogram signal, and applies a false discovery rate (FDR) method to the calculated brain network value to obtain a significant difference value of each node for a frequency band; and a brain network analysis unit which acquires node information corresponding to the disorder part of the brain according to the significant difference value of each node and classifies the degree of seriousness of a major depressive disorder (MDD) according to the number of nodes corresponding to the acquired disorder part. The apparatus can calculate a network value by using brain waves measured in each of a plurality of frequency bands, and can detect a part where an abnormal reaction has occurred by using the calculated network value, thereby enabling understanding the brain mechanism of all mental illnesses including the MDD. | en | Measurement apparatus of altered cortical region usingfunctional network and method thereof | 64039227_KR | 61536425_,75315028_,70678687_ | A61B 5/316,A61B 5/369,A61B 5/7275 | [
"A61B 5/369",
"A61B 5/24",
"A61B 5/00"
] | 148,764 |
559,325,918 | 2020-09-10 | 72,111,531 | N | A power system based on cooperative interaction between diverse users and a power grid comprises a power service module and a demand response module. The power service module is used by an electricity user or an electricity provider to query power consumption statistical data and to perform power consumption data analysis according to current power consumption statistical data. The demand response module is used to provide a demand response management function to the diverse users, and comprises a user behavior profile sub-module. The behavior profile sub-module for diverse users is used to ascertain, on the basis of a large volume of user data from different industries, diverse user behavior patterns in terms of temperature sensitivity, electricity price sensitivity, and power consumption stability, to establish and quantify separate feature indicators and perform a comprehensive assessment, to form diverse user power consumption behavior tags, and to establish multi-property power consumption behavior profiles that quantify, categorize, and combine the diverse user power consumption behavior tags. The invention aims to help electricity users or electricity providers improve commercial grid service capabilities while improving practical resource allocation, thereby facilitating energy saving and emission reduction. | en | POWER SYSTEM BASED ON COOPERATIVE INTERACTION BETWEEN DIVERSE USERS AND POWER GRID | 80733822_CN | 63976445_CN,63728844_CN,80728712_CN,68063053_CN,82808547_CN,71546140_CN,81188936_CN,82729456_CN,83051339_CN | G06F 16/245,G06F 16/2462,G06F 16/2465,G06F 16/248,G06N 3/006,G06N 3/0445,G06N 3/0454,G06N 20/10,G06Q 10/04,G06Q 10/06315,G06Q 10/067,G06Q 30/0201,G06Q 50/06,Y02D 10/00,Y04S 50/14 | [
"G06Q 30/02"
] | 159,067 |
45,606,736 | 2001-02-09 | 23,991,411 | N | Data consistent with autoimmune disease being caused by Epstein-Barr virus are shown. Based on this evidence, an effective vaccine would prevent the autoimmune disease in those vaccinated, modified or administered so that the vaccine is not itself capable of inducing autoimmune disease. In the case of anti-Sm, structures to be avoided in an Epstein-Barr virus-derived vaccine have been identified. Differences have been identified in the immune responses to Epstein-Barr infection between individuals who develop a specific autoimmune disease and those who do not. These differences are used to distinguish those who are at greater risk for developing specific autoimmune diseases from those who are a lesser risk. Assuming Epstein-Barr virus causes autoimmune disease and that Epstein-Barr virus remains latent in the patient for life, reactivation of the virus from the latent state is important in generating or maintaining the autoimmune response that culminates in autoimmune disease. Cells infected with latent virus may also encourage autoimmunity. Based on the understanding that reactivation or latency are important to produce or sustain autoimmunity, then therapies directed against Epstein-Barr virus will also be effective therapies for the autoimmune manifestations of disease for which Esptein-Barr virus is responsible. | en | DIAGNOSTICS AND THERAPY OF EPSTEIN-BARR VIRUS IN AUTOIMMUNE DISORDERS | 8498615_US | 13435261_US,37014040_US,37014041_US | A61K 39/00,A61K 39/12,A61K 39/245,A61K2039/57,A61P 37/02,C07K 14/005,C12N2710/16222,C12N2710/16234,Y02A 50/30,Y10S 435/975 | [
"C07K 14/05",
"A61K 39/245",
"A61K 39/00",
"A61P 37/02"
] | 29,916 |
50,179,903 | 1970-04-13 | 21,838,746 | Y | The disclosure describes a psychesthetic device for attachment to a television screen or tube face for the production of moving variously colored light patterns for direct or indirect visual observation. The device includes, in one embodiment, a panel of essentially opaque material having a plurality of spaced projections with thin-walled bottoms on the outside with or without a layer of irregular shaped multi-colored translucent panels on the inside of the panel. In one embodiment the panel is pigmented plastic so as to transmit only high levels of light intensity except at the ocular end of the projection where it is thinned in the molding process so as to relatively translucent or transparent to low levels of light intensity. Colored light from the television tube, is transmitted through the projections and the color pattern of the transmitted light is modified and transformed by the projections into an ever-changing light pattern with the ocular end of each projection acting as a picture element or point of light diffusion in the pattern. With the room lights and the television sound turned off, the psychesthetic light pattern is produced. The layer of multi-colored translucent panels or colored oculars are used in the combination with either colored or black and white television sets to produce similar effects. | en | PSYCHEDELIC DEVICE ATTACHABLE TO FRONT OF TELEVISION TUBE | 38768486_ | 38768486_ | H04N 11/00 | [
"H04N 11/00"
] | 39,663 |
457,491,857 | 2015-05-06 | 53,066,129 | N | An ultrasound deep brain stimulation (DBS) method and system, the ultrasound DBS method comprising: conducting medical imaging on the head of animal or human being, and generating image data; according to the image data, establishing a three-dimensional digital model of the head; according to a structure, a density, and acoustic parameter information of an ultrasound transducer array (607), establishing a three-dimensional digital model of the ultrasound transducer array (607); according to the three-dimensional digital model of the head, the three-dimensional digital model of the ultrasound transducer array (607), the structure, density, and acoustic parameters of skull and brain tissue, and the structure, density, and acoustic parameter of the ultrasound transducer array (607), generating a first ultrasound transmission sequence; controlling the ultrasound transducer array (607) to transmit an ultrasonic wave according to the first ultrasound transmission sequence, and implementing the ultrasound DBS with respect to a brain nerve nucleus to be stimulated. In the method and system, ultrasound noninvasively penetrates the skull and is focused on a deep brain region. Different ultrasound transmission sequences can realize ultrasound neural regulation and research into the mechanism of action thereof can be performed. | en | ULTRASOUND DEEP BRAIN STIMULATION METHOD AND SYSTEM | 66824291_CN | 67724421_CN,64300729_CN,66829661_CN,64130513_CN,67701646_CN,69043818_CN | A61B 5/0036,A61B 5/0042,A61B 5/055,A61B 5/24,A61B 5/377,A61B 5/7239,A61B2562/0209,A61B2562/046,A61N 7/02,A61N2007/0026,A61N2007/0052,A61N2007/0073,A61N2007/0082,G16H 50/50,G16Z 99/00 | [
"A61N 7/02",
"A61B 5/055"
] | 101,889 |
550,178,526 | 2020-10-05 | 75,738,346 | Y | An AI-based question and answer system and method can present descriptive answers to various questions of the user through the steps of user inquiry, query type classification, question-related document selection, correct answer location search, descriptive answer generation, and answer presentation steps. More specifically, in the question answering system for presenting an AI answer to a user query, a memory for storing instructions, and executing the instructions: classify a user query type of a user query based on a query type classification model, and converts a user query type to a fact query based on the type conversion model, selects a query-related document from the search target documents based on the similarity between the document vector of each of the search target documents and the fact query query vector, and understands the machine document Based on the machine reading comprehension (MRC) model, the correct answer position is searched inside the document related to the question, and the correct answer derived from the correct answer position is post-processed according to the question type to generate a short answer type answer to the user's query, and a processor configured to present an AI answer by converting a short-answer type answer into an open-ended answer based on the improved Seq2Seq model. | en | AI AI BASED QUESTION AND ANSWER SYSTEM AND METHOD | 80997107_KR | 79852985_,79940210_ | G06F 16/3329,G06F 16/3344,G06F 16/35,G06F 40/20,G06F 40/40,G06N 3/08,G06N 5/04 | [
"G06F 40/20",
"G06F 16/332",
"G06F 16/35",
"G06F 40/40",
"G06N 5/04",
"G06N 3/08"
] | 152,923 |
574,738,026 | 2021-12-14 | 74,922,809 | N | A symptom, blood data and medical image fused method for comprehensively recognizing various suspected diseases, the method using an artificial intelligence theory and intelligent recognition and intelligent analysis techniques. Symptoms, blood data and medical images are fused, and multi-data association is used for determining the relationship between data and disease recognition and a recognition rate. A disease feature item dynamically input by using a simulated annealing method is used as an input layer, an outer loop is set as a disease feature input item, an inner-layer loop is set as a feature corresponding to diseases and is used as an input item, and an improved deep neural network method is used for adjusting parameters and intelligently recognizing diseases. The dynamic input to an improved neural network framework is achieved, parameters are adaptively adjusted for disease recognition by means of interlayer calculations, and the efficiency of intelligent disease recognition is improved. By means of an artificial intelligence method, the problems of difficult multi-disease recognition, missing of other suspected diseases, etc. are solved, and comprehensive recognition of various suspected diseases becomes a key and is now used in specialized hospital departments and physical examination medical centers. | en | SYMPTOM, BLOOD DATA AND MEDICAL IMAGE FUSED METHOD FOR COMPREHENSIVELY RECOGNIZING VARIOUS SUSPECTED DISEASES | 85619044_CN,85334988_CN,70901134_CN | 70901134_CN,85334988_CN,85619044_CN | G06K 9/629,G06N 3/0454,G06N 3/08,G06T 7/0012,G06T2207/20081,G06T2207/20084,G06T2207/30024,G06V2201/03,G16H 50/20 | [
"A61B 5/145",
"G16H 50/20"
] | 167,080 |
556,831,825 | 2021-07-08 | 73,947,305 | N | The invention discloses a deep learning network architecture search method for an EEG signal classification task. The method comprises the following steps: collecting a certain amount of electroencephalogram signals of a user according to a specific brain-computer interface task, carrying out downsampling on the collected electroencephalogram signals, and then carrying out band-pass filtering; searching a deep learning network architecture by utilizing the processed electroencephalogram signals; and setting network architecture parameters, calculating optimal model internal learnable parameters on a training set, fixing the model structure to be unchanged, obtaining the internal learnable parameters by using a gradient descent method, and outputting a final network structure. The method has the advantages of being convenient to integrate and low in integration cost, a network architecture needs to be artificially designed in advance in classical deep learning-based electroencephalogram interface system development, the method can change the artificial design process, and automatic design is completed by using a deep learning network, so the development process of the existing brain-computer interface system does not need to be greatly modified, and the system can be conveniently integrated into the existing system. | en | Deep learning network architecture search method for EEG signal classification task | 70489360_,75579094_ | 60692843_,65075408_,61540236_,60624231_,70574549_,62177686_,60616563_ | G06K 9/00503,G06K 9/00536,G06K 9/6256,G06N 3/0454 | [
"G06N 3/04",
"G06K 9/00",
"G06K 9/62"
] | 157,420 |
551,969,552 | 2021-03-03 | 76,247,379 | N | The invention discloses a semi-pairing multi-view neighborhood correlation analysis method. The method comprises the following steps: 1) calculating a similarity matrix of samples in views by using a k neighborhood and a radial basis function; 2) calculating a similarity matrix of samples among different views by utilizing a similarity matrix of paired samples shared by the samples among the views and the samples in the views; 3) constructing an optimization model of semi-pairing multi-view neighborhood correlation analysis, and converting the optimization model into a generalized eigenvalue problem by using a Lagrange multiplier method; 4) solving a generalized eigenvalue problem; 5) using the projection matrix of each view to carry out dimensionality reduction on the training sample and the test sample to obtain low-dimensional representations of the training sample and the test sample; and 6) identifying the test data after dimension reduction by using a k-nearest neighbor classifier, and calculating an identification rate. Not only can the correlation between paired multi-view samples be revealed, but also the information contained in a large number of unpaired samples can be effectively utilized, so that the over-fitting phenomenon caused by the limited number of the paired samples can be effectively relieved. | en | Semi-pairing multi-view neighborhood correlation analysis method | 60425290_ | 59274137_,61155894_,67403275_,82775516_,73166130_,65524897_ | G06K 9/6215,G06K 9/6256 | [
"G06K 9/62"
] | 154,242 |
53,999,570 | 2004-07-09 | 27,357,518 | Y | A method for treating a patient with an obsession, a compulsion, an anxiety or a depression disorder, comprising applying chronic electrical stimulation to at least a portion of the patient's anterior limb of the internal capsule under conditions effective to provide the patient with at least a partial relief from obsession and/or compulsion and/or anxiety and/or depression, by means of an electrical signal generator and at least an implantable electrode having a proximal end coupled to the said signal generator and a stimulation end capable of applying said chronic electrical stimulation, wherein the distance between the two outer contacts of the electrode corresponds to the depth of the internal capsule. A method is disclosed for treating a patient with an obsession, a compulsion, an anxiety or a depression disorder. The method comprises applying electrical stimulation to at least a portion of the patient's anterior limb of the internal capsule or its surrounding structures including but not limited to the head of the caudate nucleus, putamen and nucleus accumbens. The electrical stimulation is applied by means of an electrical signal generator and at least an implantable electrode having a proximal end coupled to the signal generator and a stimulation end capable of applying electrical stimulation to the tissue. | en | Electrical stimulation of structures within the brain | 10373224_SE,5210158_US,5276141_NL,10373225_SE,5276140_BE,5276142_BE,10373223_BE,10373226_SE | 10373223_BE,5276142_BE,10373224_SE,5276141_NL,5276140_BE,10373225_SE | A61N 1/36082,A61N 1/36096,A61N 1/36153,A61N 1/36171 | [
"A61N 1/00"
] | 47,784 |
41,061,926 | 2007-01-22 | 38,501,767 | Y | A method for analyzing the interaction between protein and protein is provided to rapidly and conveniently analyze in vivo protein-protein interaction by using a vector which is able to perform bimolecular fluorescence complementation(BiFC) as well as portion-specific recombination. The method comprises the steps of: (a) introducing a nucleic acid coding a target protein or a nucleic acid coding a protein to be analyzed respectively into a vector expressing in a fusion protein form fused with an N-terminal region of a fluorescent protein and a vector expressing in a fusion protein form fused with a C-terminal region of a fluorescent protein, wherein the vector includes a nuclear localization signal at the lower portion of the promoter, and portion-specific recombination including a DNA recombination sequence pair for DNA sequence exchange occurs in the vector; (b) transforming the vector expressing the target protein and the vector expressing the protein to be analyzed into a cell and culturing it; and (c) measuring the interaction between the target protein and the protein to be analyzed through the fluorescence of the fluorescent protein, wherein the fluorescent protein is yellow fluorescent protein(YFP), green fluorescent protein(GFP), enhanced GPF, blue fluorescent protein(BFP) or cyan fluorescent protein(CFP). | en | Method for analyzing protein-protein interaction | 7628547_KR | 24619118_KR,33132538_KR | C07K2319/00,C12Q 1/00,G01N 21/6428,G01N 33/6845 | [
"C12Q 1/00",
"G01N 33/533",
"G01N 33/53",
"G01N 33/68"
] | 26,687 |
539,990,431 | 2020-03-27 | 72,941,524 | N | A photoelectric detector. Laser light emitted from a laser (2) of the detector is focused on a sample to be detected (6) by means of an objective lens of an upright fluorescence microscope (1) to excite said sample (6) to produce a fluorescence signal; an image acquisition device (3) collects the fluorescence signal output by the upright fluorescence microscope (1) and images the fluorescence signal; an input end of a first probe (4) and an input end of a second probe (5) are respectively connected to electrodes on both ends of said sample (6); an output end of the first probe (4) is electrically connected to an input end of an electric signal amplifier (7); an output end of the second probe (5) is electrically connected to an output end of a signal acquisition device (8); an output end of the electric signal amplifier (7) is electrically connected to an input end of the signal acquisition device (8); a signal processing device is used for obtaining an electric signal collected by the signal acquisition device (8), and generating and outputting electric characteristic data; and the signal processing device is further used for triggering an optical system and an electric system to operate synchronously. The photoelectric detector can reduce omission of important information in the biophysical process of said sample. | en | PHOTOELECTRIC DETECTOR | 67868160_CN | 79088903_CN,68583143_CN,79300084_CN | G01N 21/64,G01N 21/6402,G01N 21/6458,G01N2021/6463,G01N2021/6495 | [
"G01N 21/64"
] | 146,246 |
363,975,340 | 2009-01-15 | 40,956,795 | N | A cerebral blood flow quantification device that comprises: - a nuclear medicine image acquisition unit for acquiring cephalic nuclear medicine images of a subject having been administered with a radiopharmaceutical; - a count calculation unit for calculating the number of counts obtained from unit areas showing the same site in the cerebral tissue in multiple cephalic nuclear medicine images that are continuously imaged between a first time point and a second time point; - a cerebral blood flow rate calculation unit for computing, for each of the unit areas of the cerebral tissue, the cerebral blood flow rate per unit area that makes the value of a function integrated from the first time to the second time equal to the number of counts calculated by the count calculation unit, the function including cerebral blood flow rate and time elapsed from the administration of the radiopharmaceutical as variables and expressing SPECT or PET counts corresponding to the concentration of the radiopharmaceutical in brain tissue; and - a cerebral blood flow map-forming unit for forming a map showing cerebral blood flow based on the cerebral blood flow rates in the individual unit areas. The cerebral blood flow can be quantified at a high accuracy even in a relatively early stage after the administration of the radioactive drug. | en | CEREBRAL BLOOD FLOW QUANTIFICATION DEVICE, CEREBRAL BLOOD FLOW QUANTIFICATION METHOD AND PROGRAM | 6792156_ | 43681962_,43681963_,42379542_,29327168_ | A61B 5/02755,A61B 6/037,A61B 6/507,G06T 7/0012,G06T2207/10104,G06T2207/10108,G06T2207/30016 | [
"A61B 6/00"
] | 73,884 |
54,428,435 | 1992-11-27 | 25,529,303 | Y | A patient's cardiac cycle is monitored (34) for a characteristic point (54) of the cardiac cycle. Following or in response to the characteristic point, a series of field echo sequences (FIG. 3 ) are applied to generate magnetic resonance echoes (50a, 50b, 50c, etc.) at about 10 millisecond intervals (TR=10 ms) following the characteristic point. The echoes are phase encoded (44) such that echoes phase encoded in a lowest frequency or central most segment (I) of k-space are generated at regular intervals. Between temporal consecutive segment (I) echoes, echoes with higher frequency phase encoding from segments (II) and (III) (FIG. 4 ), segments (II-IV) (FIGS. 5A, 5B), etc. are generated. The views are sorted (60) into data sets (62a, 62b, 62c, etc.) such that the central most views are sent to only a single data set and at least some of the higher frequency views are conveyed to two data sets, i.e., the high frequency views are shared or commonly used by two data sets. Each data set is reconstructed (64) into a corresponding frame image representation (66). In this manner, the central most views which carry the most information are unique to each image, but higher frequency views are shared by two temporally adjacent images. The images are displayed sequentially or in another selected order on a video monitor (76). | en | Shared excitation phase encode grouping for improved throughput cardiac gated MRI cine imaging | 5408134_US | 9974854_US | A61B 5/0263,A61B 5/055,G01R 33/56,G01R 33/561,G01R 33/5617,G01R 33/5619,G01R 33/563,G01R 33/56308,G01R 33/5673 | [
"A61B 5/026",
"G01R 33/561",
"A61B 5/055",
"G01R 33/563",
"G01R 33/56",
"G01R 33/567"
] | 49,221 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.