DermaSeva

Why This Exists
India has over one million ASHA and Anganwadi workers. They are not doctors. They are community health workers, the only connection between a remote village and any medical advice. They treat paediatric pneumonia, early leprosy, severe malnutrition and fungal infections without diagnostic tools, without a specialist to call, and often without an internet signal.

The problem is not that good AI doesn't exist. The problem is it was never built for this environment: cheap hardware, no connectivity, six languages, and users that want answers, not prompts.

DermaSeva is our attempt to close that gap.

What It Actually Does
A health worker opens the app, selects a category (Skin, Child Health, or Malnutrition), and describes what they are seeing by voice, in their own language. Within seconds, they receive a structured action plan: what to do, what not to do, and whether to refer the patient to a doctor immediately.

Everything from speech recognition, medical knowledge retrieval and LLM inference happens on the device. No internet required. No data leaves the phone.

35+ conditions are covered across three domains:

Skin: ringworm, leprosy, cellulitis, eczema, impetigo, psoriasis, vitiligo, fungal infections, boils
Child health: pneumonia, dengue, tuberculosis suspect, malaria, newborn danger signs, conjunctivitis, ear infections, diarrhea with dehydration
Malnutrition: SAM, MAM, iron deficiency anemia, vitamin A/D deficiency, iodine deficiency goiter
All conditions and treatment steps are mapped directly from NHM, WHO, and IMNCI guidelines. The app does not invent.