Medical AI Assistant

The application positions itself as an AI medical assistant offering 'intelligent healthcare solutions' in a single web interface, covering consultation, monitoring, screening and emergency support.

A single-page web interface with a module for each capability: Chat, Voice, Symptoms, X-Ray/MRI, Wearable, Mental Health, Medications, 2nd Opinion, Emergency, Nutrition, Drugs, Records and Risk.

According to the app, language and vision features are powered by Groq AI and free APIs; voice transcription uses Groq Whisper, replies can be read aloud with browser text-to-speech, and image analysis uses a Groq Llama Vision model.

Smartwatch connection uses the Web Bluetooth API (Chrome/Edge only), medication reminders use the browser Notification API, and the hospital locator uses OpenStreetMap data with no API key.

Deployed on Render at medical-assistant-nannu.onrender.com. The page states that uploaded health records are processed in memory and not stored permanently.

Source code is not linked from this portfolio until a repository is verified.

No accuracy figures, benchmarks, user statistics or clinical validation are reported. None have been verified for publication.

The app states that AI image analysis is informational only and must be reviewed by a qualified radiologist; that the second-opinion feature does not replace a physician; and that pill identification is for reference and should be confirmed with a pharmacist.

Wearable connection depends on Web Bluetooth, which is available only in Chrome and Edge. This project is not a clinically validated diagnostic system and has no medical approval.