Spaces:
Sleeping
Sleeping
File size: 7,190 Bytes
2929135 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 |
# src/tools/quality_tools.py
from typing import Dict, List, Optional, Any
from typing_extensions import TypedDict # If using TypedDict
from langchain_core.tools import tool
from datetime import datetime, timedelta
from ..utils.logger import setup_logger
logger = setup_logger(__name__)
class QualityTools:
@tool
def analyze_patient_satisfaction(
self,
satisfaction_scores: List[float],
feedback_comments: List[str],
department: Optional[str] = None
) -> Dict:
"""Analyze patient satisfaction scores and feedback"""
try:
analysis = {
"metrics": {
"average_score": sum(satisfaction_scores) / len(satisfaction_scores),
"total_responses": len(satisfaction_scores),
"score_distribution": {},
"trend": "stable"
},
"feedback_analysis": {
"positive_themes": [],
"negative_themes": [],
"improvement_areas": []
},
"recommendations": []
}
# Analyze score distribution
for score in satisfaction_scores:
category = int(score)
analysis["metrics"]["score_distribution"][category] = \
analysis["metrics"]["score_distribution"].get(category, 0) + 1
# Basic sentiment analysis of feedback
positive_keywords = ["great", "excellent", "good", "satisfied", "helpful"]
negative_keywords = ["poor", "bad", "slow", "unhappy", "dissatisfied"]
for comment in feedback_comments:
comment_lower = comment.lower()
# Analyze positive feedback
for keyword in positive_keywords:
if keyword in comment_lower:
analysis["feedback_analysis"]["positive_themes"].append(keyword)
# Analyze negative feedback
for keyword in negative_keywords:
if keyword in comment_lower:
analysis["feedback_analysis"]["negative_themes"].append(keyword)
# Generate recommendations
if analysis["metrics"]["average_score"] < 7.0:
analysis["recommendations"].append("Implement immediate satisfaction improvement plan")
return analysis
except Exception as e:
logger.error(f"Error analyzing patient satisfaction: {str(e)}")
raise
@tool
def monitor_clinical_outcomes(
self,
outcomes_data: List[Dict],
benchmark_metrics: Dict[str, float]
) -> Dict:
"""Monitor and analyze clinical outcomes against benchmarks"""
try:
analysis = {
"outcome_metrics": {},
"benchmark_comparison": {},
"critical_deviations": [],
"success_areas": []
}
# Analyze outcomes by category
for outcome in outcomes_data:
category = outcome["category"]
if category not in analysis["outcome_metrics"]:
analysis["outcome_metrics"][category] = {
"success_rate": 0,
"complication_rate": 0,
"readmission_rate": 0,
"total_cases": 0
}
# Update metrics
metrics = analysis["outcome_metrics"][category]
metrics["total_cases"] += 1
metrics["success_rate"] = (metrics["success_rate"] * (metrics["total_cases"] - 1) +
outcome["success"]) / metrics["total_cases"]
# Compare with benchmarks
if category in benchmark_metrics:
benchmark = benchmark_metrics[category]
deviation = metrics["success_rate"] - benchmark
if deviation < -0.1: # More than 10% below benchmark
analysis["critical_deviations"].append({
"category": category,
"deviation": deviation,
"current_rate": metrics["success_rate"],
"benchmark": benchmark
})
elif deviation > 0.05: # More than 5% above benchmark
analysis["success_areas"].append({
"category": category,
"improvement": deviation,
"current_rate": metrics["success_rate"]
})
return analysis
except Exception as e:
logger.error(f"Error monitoring clinical outcomes: {str(e)}")
raise
@tool
def track_compliance_metrics(
self,
compliance_data: List[Dict],
audit_period: str
) -> Dict:
"""Track and analyze compliance with medical standards and regulations"""
try:
analysis = {
"compliance_rate": 0,
"violations": [],
"risk_areas": [],
"audit_summary": {
"period": audit_period,
"total_checks": len(compliance_data),
"passed_checks": 0,
"failed_checks": 0
}
}
# Analyze compliance checks
for check in compliance_data:
if check["compliant"]:
analysis["audit_summary"]["passed_checks"] += 1
else:
analysis["audit_summary"]["failed_checks"] += 1
analysis["violations"].append({
"standard": check["standard"],
"severity": check["severity"],
"date": check["date"]
})
# Identify risk areas
if check["severity"] == "high" or check.get("repeat_violation", False):
analysis["risk_areas"].append({
"area": check["standard"],
"risk_level": "high",
"recommendations": ["Immediate action required",
"Staff training needed"]
})
# Calculate overall compliance rate
total_checks = analysis["audit_summary"]["total_checks"]
if total_checks > 0:
analysis["compliance_rate"] = (analysis["audit_summary"]["passed_checks"] /
total_checks * 100)
return analysis
except Exception as e:
logger.error(f"Error tracking compliance metrics: {str(e)}")
raise# quality_tools implementation
|