Chronic Disease Copilot for Patients An AI-powered health assistant designed to help patients manage chronic conditions with voice-first interactions, intelligent symptom logging, and privacy-focused data handling. 🚀 Overview The Chronic Disease Copilot is a comprehensive full-stack application that acts as a 24/7 personal health companion. It leverages ElevenLabs for natural voice interactions and OpenAI for intelligent health analysis, all while ensuring user privacy through a "Zero Retention Mode" for HIPAA compliance. Whether you need to log symptoms via voice, interpret complex lab results, or prepare for an upcoming doctor's appointment, this copilot streamlines the process with empathy and precision. The platform uniquely identifies "weak" health risk signals by analyzing real-world, self-reported data and everyday inputs described by a person (not formal medical records), making preventive care accessible and actionable for everyone.
✨ Key Features
🎙️ Voice-Activated Symptom Logger
Log symptoms naturally using your voice. AI analyzes speech to classify severity and detect patterns. Returns empathetic audio feedback using ElevenLabs.
🔒 Zero Retention Mode (HIPAA Compliant)
Toggle privacy mode to ensure sensitive health data is processed but never stored. Perfect for users who prioritize absolute data privacy.
🧪 Smart Lab Results Interpretation
Upload or view lab results with AI-powered plain-language explanations. Visual trend analysis (e.g., "HbA1c improved by 11%"). Audio narration of results for accessibility.
📅 Appointment Preparation Coach
Generates comprehensive summaries for upcoming doctor visits. Suggests personalized questions based on your recent symptoms and logs. Creates a downloadable PDF brief to share with your provider.
🔐 Secure Authentication & Profile
Multi-factor authentication (MFA) flow. Comprehensive patient profile management (Conditions, Medications, Allergies).
🛠️ Technology Stack Frontend
React 19 with Vite for lightning-fast development. Tailwind CSS for modern, responsive styling. Framer Motion for smooth, organic animations. Lucide React for beautiful iconography. Web Speech API for native browser speech recognition.
Backend
Node.js & Express for the API server. Better-SQLite3 for robust local data storage. OpenAI API for symptom analysis and medical insights. ElevenLabs API for unparalleled text-to-speech realism.
🚀 Getting Started Follow these steps to set up the project locally. Prerequisites
Node.js (v18 or higher) npm or yarn
Installation
Clone the repository:
bashgit clone https://github.com/your-username/chronic-disease-copilot.git cd chronic-disease-copilot
📱 Usage Guide
Dashboard: The central hub for your health metrics, daily tasks, and quick actions. Symptom Logging: Click the microphone icon to speak your symptoms. Valid categories include pain, nausea, dizziness, etc. Privacy Toggle: Use the switch in the dashboard to enable/disable "Zero Retention Mode". When enabled, no data is written to the database. Lab Results: Navigate to the Labs section to view trends and listen to AI explanations of your blood work.
💡 Inspiration Living with a chronic condition often feels isolating. Patients juggle medications, track symptoms in scattered notes, and struggle to communicate their daily experiences to doctors during brief appointments. We were inspired by the stories of individuals who felt unheard by the healthcare system—people who noticed subtle changes in their health but lacked the tools to document and articulate these "weak signals" before they became serious problems. We wanted to create a compassionate, always-available companion that bridges the gap between patients and their care teams.
🎯 What It Does The Chronic Disease Copilot empowers patients to:
Capture health data effortlessly through natural voice conversations, eliminating the friction of manual logging. Detect early warning signs by analyzing patterns in self-reported symptoms, mood, and daily activities—data that traditional medical systems often miss. Translate medical jargon into plain language, making lab results and clinical recommendations accessible to everyone. Prepare for medical appointments with AI-generated summaries and personalized questions that ensure productive conversations with providers. Maintain complete privacy with Zero Retention Mode, giving users full control over their sensitive health information.
🔨 How We Built It We architected the application as a privacy-first, full-stack solution:
Voice Interface: Integrated the Web Speech API for browser-based speech recognition and ElevenLabs for natural, empathetic text-to-speech responses. Intelligent Analysis: Connected OpenAI's API to process symptom descriptions, identify severity levels, and generate actionable insights from unstructured health narratives. Privacy Architecture: Implemented a dual-mode system where users can toggle between persistent storage and ephemeral processing (Zero Retention Mode) to meet HIPAA standards. Data Persistence: Used Better-SQLite3 for local, encrypted storage of health logs, medication schedules, and lab results. User Experience: Designed an intuitive React interface with Framer Motion animations and Tailwind CSS to make health tracking feel approachable rather than clinical.
🚧 Challenges We Ran Into
Balancing Privacy and Functionality: Building a system that could analyze health data deeply while offering true zero-retention mode required careful API design and state management. Voice Recognition Accuracy: Medical terminology and symptom descriptions are complex. We had to implement robust error handling and contextual prompts to improve speech-to-text accuracy. Empathetic AI Responses: Generating AI responses that felt supportive rather than robotic required extensive prompt engineering and testing with diverse health scenarios. Real-Time Audio Processing: Streaming audio responses from ElevenLabs while maintaining low latency proved technically challenging, especially for longer explanations.
🏆 Accomplishments That We're Proud Of
Privacy-First Design: Successfully implemented true Zero Retention Mode that processes sensitive data without any persistent storage. Accessible Health Tech: Created audio narration for all major features, making the app usable for visually impaired patients or those with literacy challenges. Pattern Detection: Built an AI system that identifies subtle health trends from conversational symptom logs—something even patients themselves might miss. End-to-End Workflow: Delivered a complete patient journey from symptom logging to appointment preparation, all in one seamless interface.
📚 What We Learned
Patient-Centered Design: We learned that the best health tech doesn't just store data—it actively helps patients feel understood and prepared for medical conversations. The Power of Weak Signals: By analyzing everyday, informal health descriptions, we can catch early warning signs that traditional medical systems miss because they only look at formal clinical data. Privacy as a Feature: Offering true data control isn't just about compliance—it's a competitive advantage that builds trust with users who've been burned by data breaches. Voice is the Future of Health Logging: Speaking is faster, more natural, and more accessible than typing, especially for older adults or those managing multiple conditions.
🔮 What's Next for Chronic Disease Copilot
Predictive Health Models: Train custom ML models on anonymized symptom patterns to predict flare-ups before they happen. Care Team Integration: Build secure portals where patients can share curated summaries with their doctors, physical therapists, and caregivers. Wearable Device Sync: Integrate with Apple Health, Fitbit, and other platforms to automatically pull in vitals and activity data. Medication Reminders & Tracking: Add smart reminders with voice confirmations and side effect monitoring. Community Support: Create opt-in, anonymized groups where patients with similar conditions can share experiences and coping strategies. Multi-Language Support: Expand beyond English to serve diverse patient populations globally.
🤝 Contributing Contributions are welcome! Please feel free to submit a Pull Request.
Fork the project Create your feature branch (git checkout -b feature/AmazingFeature) Commit your changes (git commit -m 'Add some AmazingFeature') Push to the branch (git push origin feature/AmazingFeature) Open a Pull Request
📄 License This project is licensed under the MIT License - see the LICENSE file for details.
Built With
- css
- elevenlab
- express.js
- node.js
- react
- sqlite
Log in or sign up for Devpost to join the conversation.