A Diagnostic Assistant that captures patient–clinician conversations and turns them into structured EMR data, diagnoses, and insights.
- Login: EMT, Nurse, or Physician sign in to their respective dashboard.
- Dashboard & audio: Record clinical audio or get text input where Synapse uses Gemini API to extract vitals, symptoms, and diagnosis from the recording.
- Speech-to-Text: Convert clinician or patient interview text from a select audio or text in another language into structured transcript entries
- AI summarization: Extract vitals, symptoms, and diagnosis from audio, text, or file input.
- ICD code mapping & suggested treatments: Auto-tag conditions and recommend medications with physicist's final approval
- Patient-friendly summaries: Plain-language recovery plans and follow-up instructions.
- EMR integration: Audit ready payload for external EMR systems.
- Faster documentation during consultations with the help of AI tool
- Telehealth transcription with diagnosis and recovery plans support
- Emergency room triage documentation
- Multilingual support for documentation support
- Compliance and audit record generation
- Backend: Django, Django REST Framework, SQLite, Python
- Frontend: React, Vite, TypeScript, CSS
- AI: Gemini API (Gemini 2.5 Flash)
# From project root
python -m venv .venv
.venv\Scripts\activate # Windows
# source .venv/bin/activate # macOS/Linux
pip install -r requirements.txt
python manage.py migrate
python manage.py runserverDemo users (create with python manage.py create_demo_users):
- physician / physician123
- nurse / nurse123
- emt / emt123
Gemini audio extraction: set GEMINI_API_KEY in the environment (or in Django settings) so the /api/audio/extract/ endpoint can call the Gemini API.
For more information about endpoints, check demos.md.
cd frontend
npm install
npm run devApp runs at http://localhost:5173 and proxies /api to the Django backend.