File size: 4,283 Bytes
390cffd | 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 | from flask import Flask, request, jsonify
from flask_cors import CORS
import google.generativeai as genai
import os
import re
app = Flask(__name__)
CORS(app)
# Configure Gemini API
GEMINI_API_KEY = "AIzaSyCKKdpQ5Fuvbgjlr_tUFejUn0AWrA99BP0"
genai.configure(api_key=GEMINI_API_KEY)
# Initialize the model
model = genai.GenerativeModel('gemini-1.5-flash')
# OncoConnect knowledge base (from your document)
KNOWLEDGE_BASE = """
You are a helpful support chatbot for OncoConnect, a pathology research platform. Answer questions based on this information:
ONBOARDING & ACCOUNTS:
- Upload pathology images → get cancer/chemo-response indications and auto-matched trials
- Create account: Hit Log In → email or OAuth (Google/GitHub). Verify email
- Roles: Researcher, Clinician, Student, Admin
- Edit profile: Profile → Edit → update bio, avatar, institution, tags → Save
- Change notifications: Profile → Settings → Notifications
- Delete account: Profile → Settings → Privacy → Delete account
PATHOLOGY ANALYZER:
- File types: PNG/JPEG for demo; WSI via SVS/TIFF/OME-TIFF
- Max size: PNG/JPEG ≤ 25MB; WSI ≤ 2GB
- Anonymize slides: Remove PHI, crop labels, strip metadata
- "Cancer: Yes/No" with confidence (0-100%) - exploratory only, not diagnostic
- Saved cases: Analyzer → Saved Cases → rename/delete with ⋯ menu
CLINICAL TRIALS:
- Trials appear automatically after analysis
- Filter by phase, status, distance, biomarkers
- Demo data refreshes weekly
- Contact trials: Open trial → Site details → email/phone
- Export matched trials: Export CSV from Trials panel
SCORES & RISK:
- Risk score (0-100): Research composite risk, exploration only
- NOT a diagnostic tool - research/education only
- Model explanations via saliency heatmap
- Cite as: OncoConnect (version X), research prototype, non-diagnostic
CHALLENGES:
- View: Go to Challenges; filter by cancer type/skill
- Enroll: Open challenge → Enroll
- Submit: Upload artifact (CSV/zip/notebook)
- Teams: Create/join teams; invite collaborators
- Leaderboard shows top-10 + your position
LEADERBOARD & PROFILES:
- Monthly rankings, solved challenges, average scores, badges
- Connect with others via profiles
- Filter by specialty/cancer type
DATA & PRIVACY:
- Demo: browser local storage; Enterprise: encrypted at rest
- Private by default; can mark Public/Team
- NO PHI allowed - de-identified data only
- NOT FDA-cleared - research/education only
- Free demo tier with limits; academic access available
Keep answers concise and helpful. If unsure, suggest contacting support.
"""
def get_response(user_message):
try:
# Create the prompt with context
prompt = f"{KNOWLEDGE_BASE}\n\nUser question: {user_message}\n\nProvide a helpful, concise answer:"
# Generate response using Gemini
response = model.generate_content(prompt)
# Clean up the response
answer = response.text.strip()
# Limit response length
if len(answer) > 500:
answer = answer[:500] + "..."
return answer
except Exception as e:
print(f"Error generating response: {e}")
return "I apologize, but I'm having trouble processing your request right now. Please try asking again or contact support for assistance."
@app.route('/chat', methods=['POST'])
def chat():
try:
data = request.get_json()
user_message = data.get('message', '').strip()
if not user_message:
return jsonify({'error': 'No message provided'}), 400
# Generate response
response = get_response(user_message)
return jsonify({'response': response})
except Exception as e:
print(f"Error in chat endpoint: {e}")
return jsonify({'error': 'Internal server error'}), 500
@app.route('/health', methods=['GET'])
def health():
return jsonify({'status': 'healthy'})
if __name__ == '__main__':
print("🤖 OncoConnect Support Chatbot starting...")
print("📱 Frontend: Open any page in your browser")
print("🔗 Backend: Running on http://localhost:5000")
print("💬 Chatbot: Available on all pages as floating button")
app.run(debug=True, host='0.0.0.0', port=5000)
|