Get a daily reminder to log into Product Hunt

Free, open-source AI skin disease detection. Upload a photo, get a prediction in under 2 seconds. No signup, no data stored — images are processed in memory and discarded immediately. Two CNN models: a 7-class skin lesion classifier and a monkeypox detector (95% accuracy). Each result includes an AI-generated explanation via Groq Llama 4. Built with Flask, TensorFlow Lite, and Tailwind CSS. Fully open source. Educational use only.

Framer — Launch websites with enterprise needs at startup speeds.

Launch websites with enterprise needs at startup speeds.

Hey Product Hunt! I'm a CS student and I built SKIN as my capstone thesis project. The idea came from a simple observation: skin diseases are one of the most common health concerns, but access to dermatologists is limited in many parts of the world. I wanted to build something that gives people a starting point — not a diagnosis, but an informed nudge to see a doctor. A few things I'm proud of: - Privacy-first: images never touch disk. Processed in memory, discarded after prediction. Zero user data to leak. - Lightweight: the TFLite model is just 2.7 MB. No GPU needed. Runs on Render's free tier. - AI explanations: Groq's Llama 4 Scout generates plain-language context for each prediction, so you understand what "bcc — 73% confidence" actually means. - No signup, no paywall, no ads. Just upload and analyze. The entire codebase is open source (MIT). Would love your feedback on the UI, the model accuracy, or ideas for new conditions to add. GitHub: github.com/jsgbmn/skin-disease-app Happy to answer any questions!

Framer — Launch websites with enterprise needs at startup speeds.

Launch websites with enterprise needs at startup speeds.

MailAmp

Get a daily reminder to log into Product Hunt

Get a daily reminder to log into Product Hunt

Skin Disease Identification

Previous MailAmp Launches

Previous MailAmp Launches