SECE‑MedAssist is a hybrid healthcare AI system combining:
- MedGemma for clinical reasoning
- SECE (Simulated Emotion, Context, and Empathy) for emotional intelligence
- A Portable Knowledge Base (PKB) for offline medical knowledge
This repository contains the documentation and prototype code used for the MedGemma Impact Challenge submission.
A lightweight analog emotional‑weight engine that modulates tone, clarity, and emotional resonance in patient communication.
A prototype demonstrating how MedGemma outputs are processed and emotionally modulated by SECE.
A modular, offline medical knowledge unit inspired by high‑performance biomedical data pipelines.
To create a patient‑communication assistant that is:
- medically accurate
- emotionally supportive
- privacy‑preserving
- fully functional offline
This repository supports the Kaggle write‑up and video demonstration for the MedGemma Impact Challenge.