yarsi-ai/sicaps

# SICAPS **Smart AI System for Scabies Screening** Web-based chatbot that performs preliminary scabies screening through interactive conversation — extracting clinical keywords, calculating risk scores, and providing personalized recommendations. --- ## About SICAPS Scabies is a highly contagious skin disease commonly found in high-density environments such as Islamic boarding schools (pesantren). Limited access to medical professionals causes many cases to go undetected or receive delayed treatment. SICAPS provides an accessible preliminary screening solution. Through natural conversation with an AI chatbot, the system: 1. **Explores symptoms** adaptively (itch intensity, timing, body locations, contact history, skin lesions, risk factors) 2. **Extracts clinical keywords** from user responses using LLM 3. **Calculates risk scores** deterministically on the backend (not by AI) 4. **Generates output** including risk level (High/Moderate/Low), perception response, action recommendations, and personalized care advice This project is part of dr. Erlina Wijayanti's research at Yarsi University, targeting screening accuracy of ≥80% compared to clinical diagnosis by physicians. --- ## Features (MVP — Phase 1) - Interactive AI chat with adaptive conversation flow (not rigid Q&A) - Voice input (Speech-to-Text) + voice output (Text-to-Speech) via Web Speech API - Keyword extraction by LLM + deterministic scoring engine on backend - Questionnaire mode as fallback when AI is offline - Bilingual — Bahasa Indonesia + English - Risk level output + 4-part personalized results - PDF download of screening results (with Yarsi header) - Shareable result link - Anonymous screening (no login required) --- ## Tech Stack | Layer | Technology | | ---------- | --------------------------------------------------------- | | Frontend | Next.js 16 (App Router), React 19, Tailwind CSS, Radix UI | | Backend | Next.js API Routes …