Sympto Track AI Hybrid RAG Chatbot for Symptom Monitoring

SymptoTrackAI is a hybrid Retrieval-Augmented Generation (RAG) chatbot designed for personalized symptom monitoring and medical report analysis. Built using Flask, it enables users to engage in conversational queries via text input while uploading PDF medical reports for contextual integration. Reports are processed with PyPDF2 for text extraction, chunked, and embedded using the sentence-transformers/all-MiniLM-L6-v2 model into a persistent Chroma vector database. The system employs a RAG pipeline with Groq's Llama-3.3-70B-Versatile LLM for accurate, context-grounded responses restricted to medical domains.

Upon upload, the LLM generates an initial professional summary of key findings (e.g., patient details, vital parameters, diagnoses), which primes the conversation memory. Users can then ask follow-up questions about symptoms, diseases, precautions, or report specifics, with responses drawn exclusively from retrieved context and chat history for reliability and hallucination mitigation. Features include user authentication, multiple conversation management, automatic title generation, and conversation persistence via SQLite. SymptoTrackAI provides a secure, user-friendly tool for health tracking and informed discussions.

Keywords

SymptoTrackAI, medical chatbot, symptom monitoring, Retrieval-Augmented Generation (RAG), PDF report analysis, hybrid AI, LangChain, Chroma vector database, HuggingFace embeddings, Groq LLM, Llama-3.3-70B, Flask web application, health informatics, personalized medicine, conversational AI.