MedSpeech-012324

Sleeping

App Files Files Community

awacke1 commited on Dec 5, 2023

Commit

04d4dca

•

1 Parent(s): 9d841e8

Update app.py

Browse files

Files changed (1) hide show

app.py +37 -43

app.py CHANGED Viewed

@@ -3,6 +3,10 @@ import json
 import pandas as pd
 import streamlit.components.v1 as components
 # Function to load JSONL file into a DataFrame
 def load_jsonl(file_path):
     data = []
@@ -11,15 +15,12 @@ def load_jsonl(file_path):
             data.append(json.loads(line))
     return pd.DataFrame(data)
-# Your filter_by_keyword function
-def filter_by_keyword(data, term):
-    # Your filtering logic here
-    return data[data['column_name'].str.contains(term)]
 # Function to generate HTML with textarea
-def generate_html_with_textarea(row):
-    first_three_columns_text = ' '.join([f"{col}: {row[col]}" for col in row.index[:3]])
     return f'''
     <!DOCTYPE html>
     <html>
@@ -36,16 +37,14 @@ def generate_html_with_textarea(row):
     <body>
         <h1>🔊 Read It Aloud</h1>
         <textarea id="textArea" rows="10" cols="80">
-    {first_three_columns_text}
         </textarea>
         <br>
         <button onclick="readAloud()">🔊 Read Aloud</button>
     </body>
     </html>
     '''
-filtered_data = pd.DataFrame()
 # Streamlit App 🚀
 st.title("AI Medical Explorer with Speech Synthesis 🎙")
@@ -62,13 +61,7 @@ data = large_data if file_option == "usmle_16.2MB.jsonl" else small_data
 # Top 20 healthcare terms for USMLE
 top_20_terms = ['Heart', 'Lung', 'Pain', 'Memory', 'Kidney', 'Diabetes', 'Cancer', 'Infection', 'Virus', 'Bacteria', 'Gastrointestinal', 'Skin', 'Blood', 'Surgery']
-# Initialize session state for tracking the last clicked row
-if 'last_clicked_row' not in st.session_state:
-    st.session_state['last_clicked_row'] = None
-# Streamlit app
 with st.expander("Search by Common Terms 📚"):
     cols = st.columns(4)
     for term in top_20_terms:
@@ -76,32 +69,33 @@ with st.expander("Search by Common Terms 📚"):
             if st.button(f"{term}"):
                 filtered_data = filter_by_keyword(data, term)
                 st.write(f"Filter on '{term}' 📊")
-                # Display clickable rows in the dataframe
-                for idx, row in filtered_data.iterrows():
-                    link = f"[Row {idx}]('javascript:void(0);')"
-                    st.markdown(link, unsafe_allow_html=True)
-                # Capture the clicked row index
-                row_index = st.experimental_get_query_params().get("row_index")
-                if row_index:
-                    selected_row = filtered_data.loc[int(row_index[0])]
-                    html_content = generate_html_with_textarea(selected_row)
-                    components.html(html_content, width=1280, height=1024)
-# Always show the first row if filtered_data is not empty
-if not filtered_data.empty:
-    first_row_html = generate_html_with_textarea(filtered_data.iloc[0])
-    components.html(first_row_html, width=1280, height=1024)
-# Inject HTML5 and JavaScript for styling
-st.markdown("""
-<style>
-    .big-font {
-        font-size:24px !important;
-    }
-</style>
-""", unsafe_allow_html=True)
 # Markdown and emojis for the case presentation
 st.markdown("# 🏥 Case Study: 32-year-old Woman's Wellness Check")

 import pandas as pd
 import streamlit.components.v1 as components
+# Initialize session state for tracking the last clicked row
+if 'last_clicked_row' not in st.session_state:
+    st.session_state['last_clicked_row'] = None
 # Function to load JSONL file into a DataFrame
 def load_jsonl(file_path):
     data = []
             data.append(json.loads(line))
     return pd.DataFrame(data)
+# Function to filter DataFrame by keyword
+def filter_by_keyword(df, keyword):
+    return df[df.apply(lambda row: row.astype(str).str.contains(keyword).any(), axis=1)]
 # Function to generate HTML with textarea
+def generate_html_with_textarea(text_to_speak):
     return f'''
     <!DOCTYPE html>
     <html>
     <body>
         <h1>🔊 Read It Aloud</h1>
         <textarea id="textArea" rows="10" cols="80">
+    {text_to_speak}
         </textarea>
         <br>
         <button onclick="readAloud()">🔊 Read Aloud</button>
     </body>
     </html>
     '''
 # Streamlit App 🚀
 st.title("AI Medical Explorer with Speech Synthesis 🎙")
 # Top 20 healthcare terms for USMLE
 top_20_terms = ['Heart', 'Lung', 'Pain', 'Memory', 'Kidney', 'Diabetes', 'Cancer', 'Infection', 'Virus', 'Bacteria', 'Gastrointestinal', 'Skin', 'Blood', 'Surgery']
+# Create Expander and Columns UI for terms
 with st.expander("Search by Common Terms 📚"):
     cols = st.columns(4)
     for term in top_20_terms:
             if st.button(f"{term}"):
                 filtered_data = filter_by_keyword(data, term)
                 st.write(f"Filter on '{term}' 📊")
+                with st.sidebar:
+                    st.dataframe(filtered_data)
+                if not filtered_data.empty:
+                    html_blocks = []
+                    for idx, row in filtered_data.iterrows():
+                        question_text = row.get("question", "No question field")
+                        documentHTML5 = generate_html_with_textarea(question_text)
+                        html_blocks.append(documentHTML5)
+                    all_html = ''.join(html_blocks)
+                    components.html(all_html, width=1280, height=1024)
+# Text input for search keyword
+search_keyword = st.text_input("Or, enter a keyword to filter data:")
+if st.button("Search 🕵️‍♀️"):
+    filtered_data = filter_by_keyword(data, search_keyword)
+    st.write(f"Filtered Dataset by '{search_keyword}' 📊")
+    st.dataframe(filtered_data)
+    if not filtered_data.empty:
+        html_blocks = []
+        for idx, row in filtered_data.iterrows():
+            question_text = row.get("question", "No question field")
+            documentHTML5 = generate_html_with_textarea(question_text)
+            html_blocks.append(documentHTML5)
+        all_html = ''.join(html_blocks)
+        components.html(all_html, width=1280, height=1024)
 # Markdown and emojis for the case presentation
 st.markdown("# 🏥 Case Study: 32-year-old Woman's Wellness Check")