vedaMD / src /simple_medical_chatbot.py
sniro23's picture
Initial commit without binary files
19aaa42
"""
Simple Medical Chatbot Interface v2.0
Beautiful Gradio interface for the simplified medical RAG system
"""
import gradio as gr
import time
import json
from datetime import datetime
from typing import List, Tuple, Dict, Any
# Import our simplified medical RAG system
from simple_medical_rag import SimpleMedicalRAG, MedicalResponse
class SimpleMedicalChatbot:
"""Professional medical chatbot interface using simplified RAG system"""
def __init__(self):
"""Initialize the medical chatbot"""
self.rag_system = None
self.chat_history = []
self.session_stats = {
"queries_processed": 0,
"total_response_time": 0,
"avg_confidence": 0,
"session_start": datetime.now().strftime("%Y-%m-%d %H:%M:%S")
}
# Initialize RAG system
self._initialize_rag_system()
def _initialize_rag_system(self):
"""Initialize the RAG system"""
try:
print("πŸš€ Initializing Medical RAG System...")
self.rag_system = SimpleMedicalRAG()
print("βœ… Medical RAG System initialized successfully!")
except Exception as e:
print(f"❌ Error initializing RAG system: {e}")
self.rag_system = None
def process_query(self, query: str, history: List[Tuple[str, str]]) -> Tuple[List[Tuple[str, str]], str]:
"""Process medical query and return response"""
if not self.rag_system:
error_msg = "❌ **System Error**: Medical RAG system not initialized. Please refresh and try again."
history.append((query, error_msg))
return history, ""
if not query.strip():
return history, ""
start_time = time.time()
try:
# Process query with RAG system
response = self.rag_system.query(query, k=5)
# Format response for display
formatted_response = self._format_response_for_display(response)
# Update session statistics
query_time = time.time() - start_time
self._update_session_stats(query_time, response.confidence)
# Add to chat history
history.append((query, formatted_response))
return history, ""
except Exception as e:
error_msg = f"❌ **Error processing query**: {str(e)}\n\n⚠️ Please try rephrasing your question or contact support."
history.append((query, error_msg))
return history, ""
def _format_response_for_display(self, response: MedicalResponse) -> str:
"""Format medical response for beautiful display in Gradio"""
# Confidence level indicator
confidence_emoji = "🟒" if response.confidence > 0.7 else "🟑" if response.confidence > 0.5 else "πŸ”΄"
confidence_text = f"{confidence_emoji} **Confidence: {response.confidence:.1%}**"
# Response type indicator
type_emoji = "πŸ’Š" if "dosage" in response.response_type else "🚨" if "emergency" in response.response_type else "πŸ₯"
# Main response
formatted_response = f"""
{type_emoji} **Medical Information**
{response.answer}
---
πŸ“Š **Response Details**
{confidence_text}
πŸ“š **Sources**: {len(response.sources)} documents referenced
"""
# Add top sources
if response.sources:
formatted_response += "πŸ“– **Primary Sources**:\n"
for i, source in enumerate(response.sources[:3], 1):
doc_name = source['document'].replace('.pdf', '').replace('-', ' ').title()
formatted_response += f"{i}. {doc_name} (Relevance: {source['relevance_score']:.1%})\n"
formatted_response += "\n"
# Add medical disclaimer
formatted_response += f"""
---
{response.medical_disclaimer}
πŸ”— **Note**: This response is based on Sri Lankan maternal health guidelines and should be used in conjunction with current clinical protocols.
"""
return formatted_response
def _update_session_stats(self, query_time: float, confidence: float):
"""Update session statistics"""
self.session_stats["queries_processed"] += 1
self.session_stats["total_response_time"] += query_time
# Update average confidence
current_avg = self.session_stats["avg_confidence"]
queries = self.session_stats["queries_processed"]
self.session_stats["avg_confidence"] = ((current_avg * (queries - 1)) + confidence) / queries
def get_system_info(self) -> str:
"""Get system information for display"""
if not self.rag_system:
return "❌ **System Status**: Not initialized"
try:
stats = self.rag_system.get_system_stats()
system_info = f"""
πŸ₯ **Sri Lankan Maternal Health Assistant v2.0**
πŸ“Š **System Status**: {stats['status'].upper()} βœ…
**Knowledge Base**:
β€’ πŸ“š Total Documents: {stats['vector_store']['total_chunks']:,} medical chunks
β€’ 🧠 Embedding Model: {stats['vector_store']['embedding_model']}
β€’ πŸ’Ύ Vector Store Size: {stats['vector_store']['vector_store_size_mb']} MB
β€’ ⚑ Approach: Simplified document-based retrieval
**Content Distribution**:
"""
# Add content distribution
for content_type, count in stats['vector_store']['content_type_distribution'].items():
percentage = (count / stats['vector_store']['total_chunks']) * 100
content_info = content_type.replace('_', ' ').title()
system_info += f"β€’ {content_info}: {count:,} chunks ({percentage:.1f}%)\n"
return system_info
except Exception as e:
return f"❌ **Error retrieving system info**: {str(e)}"
def get_session_stats(self) -> str:
"""Get session statistics for display"""
if self.session_stats["queries_processed"] == 0:
return "πŸ“ˆ **Session Statistics**: No queries processed yet"
avg_response_time = self.session_stats["total_response_time"] / self.session_stats["queries_processed"]
return f"""
πŸ“ˆ **Session Statistics**
πŸ• **Session Started**: {self.session_stats["session_start"]}
πŸ“ **Queries Processed**: {self.session_stats["queries_processed"]}
⚑ **Avg Response Time**: {avg_response_time:.2f} seconds
🎯 **Avg Confidence**: {self.session_stats["avg_confidence"]:.1%}
"""
def clear_chat(self) -> Tuple[List, str]:
"""Clear chat history"""
self.chat_history = []
return [], ""
def get_example_queries(self) -> List[str]:
"""Get example medical queries"""
return [
"What is the dosage of magnesium sulfate for preeclampsia?",
"How to manage postpartum hemorrhage emergency?",
"Normal fetal heart rate during labor monitoring?",
"Management protocol for breech delivery?",
"Antenatal care schedule for high-risk pregnancies?",
"Signs and symptoms of preeclampsia?",
"When to perform cesarean delivery?",
"Postpartum care guidelines for new mothers?"
]
def create_medical_chatbot_interface():
"""Create the main Gradio interface"""
# Initialize chatbot
chatbot = SimpleMedicalChatbot()
# Custom CSS for medical theme
css = """
.gradio-container {
background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
}
.medical-header {
background: white;
padding: 20px;
border-radius: 10px;
margin-bottom: 20px;
box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
}
.chat-container {
background: white;
border-radius: 15px;
box-shadow: 0 8px 25px rgba(0, 0, 0, 0.1);
}
.medical-disclaimer {
background: #fff3cd;
border: 1px solid #ffeaa7;
border-radius: 8px;
padding: 15px;
margin: 10px 0;
color: #856404;
}
.example-queries {
background: #e8f5e8;
border-radius: 8px;
padding: 15px;
margin: 10px 0;
}
"""
with gr.Blocks(css=css, title="Sri Lankan Maternal Health Assistant", theme=gr.themes.Soft()) as interface:
# Header
gr.Markdown("""
# πŸ₯ Sri Lankan Maternal Health Assistant v2.0
### Simplified Document-Based Medical RAG System
**Professional medical guidance based on Sri Lankan maternal health guidelines**
""", elem_classes=["medical-header"])
with gr.Row():
with gr.Column(scale=2):
# Main chat interface
with gr.Group(elem_classes=["chat-container"]):
gr.Markdown("## πŸ’¬ Medical Query Interface")
chatbot_display = gr.Chatbot(
label="Medical Assistant",
height=500,
show_label=False,
container=True,
bubble_full_width=False
)
with gr.Row():
query_input = gr.Textbox(
placeholder="Ask a medical question about maternal health...",
label="Your Medical Query",
lines=2,
scale=4
)
submit_btn = gr.Button("πŸ” Ask", variant="primary", scale=1)
with gr.Row():
clear_btn = gr.Button("πŸ—‘οΈ Clear Chat", variant="secondary")
refresh_btn = gr.Button("πŸ”„ Refresh System", variant="secondary")
with gr.Column(scale=1):
# System information and examples
with gr.Group():
gr.Markdown("## πŸ“Š System Information")
system_info_display = gr.Markdown(
chatbot.get_system_info(),
label="System Status"
)
with gr.Group():
gr.Markdown("## πŸ“ˆ Session Statistics")
session_stats_display = gr.Markdown(
chatbot.get_session_stats(),
label="Current Session"
)
# Example queries
with gr.Group(elem_classes=["example-queries"]):
gr.Markdown("## πŸ’‘ Example Queries")
example_queries = chatbot.get_example_queries()
for i, example in enumerate(example_queries[:4]):
example_btn = gr.Button(
f"πŸ“ {example}",
variant="secondary",
size="sm"
)
example_btn.click(
fn=lambda x=example: x,
outputs=query_input
)
# Medical disclaimer
gr.Markdown("""
## ⚠️ Important Medical Disclaimer
This system provides information from Sri Lankan maternal health guidelines for **educational purposes only**.
**Always consult qualified healthcare providers for**:
- Medical decisions and patient care
- Emergency medical situations
- Clinical diagnosis and treatment
- Medication administration
This tool is designed to **supplement**, not replace, professional medical judgment.
""", elem_classes=["medical-disclaimer"])
# Event handlers
def submit_query(query, history):
"""Handle query submission"""
new_history, _ = chatbot.process_query(query, history)
return new_history, "", chatbot.get_session_stats()
def refresh_system():
"""Refresh system information"""
return chatbot.get_system_info(), chatbot.get_session_stats()
def clear_chat_handler():
"""Handle chat clearing"""
new_history, _ = chatbot.clear_chat()
return new_history, "", chatbot.get_session_stats()
# Connect event handlers
submit_btn.click(
fn=submit_query,
inputs=[query_input, chatbot_display],
outputs=[chatbot_display, query_input, session_stats_display]
)
query_input.submit(
fn=submit_query,
inputs=[query_input, chatbot_display],
outputs=[chatbot_display, query_input, session_stats_display]
)
clear_btn.click(
fn=clear_chat_handler,
inputs=[],
outputs=[chatbot_display, query_input, session_stats_display]
)
refresh_btn.click(
fn=refresh_system,
inputs=[],
outputs=[system_info_display, session_stats_display]
)
return interface
def main():
"""Main function to launch the medical chatbot"""
print("πŸš€ Launching Sri Lankan Maternal Health Assistant v2.0")
print("=" * 60)
# Create and launch interface
interface = create_medical_chatbot_interface()
# Launch with custom settings
interface.launch(
server_name="0.0.0.0",
server_port=7860,
share=True, # Enable public sharing
show_error=True,
inbrowser=True,
debug=True
)
if __name__ == "__main__":
main()