Inspiration
Medical lab reports are filled with complex clinical shorthand and dense columns of numbers. Most patients cannot interpret their own results easily, which leads to unnecessary anxiety, confusion, and a massive health literacy gap. We wanted to build a solution that empowers patients to understand their health instantly.
What it does
KleckAdvice Agent is an AI-powered assistant that translates complex medical reports into intuitive, patient-friendly insights. When a user uploads a report, the system processes it using OCR and passes it to an intelligent LangGraph workflow. The agent flags abnormal metrics, generates a simplified health risk score, and maps the data into clear, interactive visual dashboards. It also features a secure, contextual Q&A chat box where patients can ask specific follow-up questions directly grounded in their report data.
Challenges we ran into
One of our biggest hurdles was creating a pipeline robust enough to handle various unstructured lab report layouts without failing. Standard rule-based parsing breaks easily on unexpected formats. We solved this by implementing an LLM-driven adaptive extraction pipeline that dynamically maps diverse document text into structured, clean data.
Accomplishments that we're proud of
We are incredibly proud of successfully building a working multi-language microservices architecture under tight hackathon timelines—connecting a Spring Boot entry proxy smoothly with a specialized FastAPI AI worker engine to safely route and process binary data streams.
What's next for KleckAdvice
We plan to implement full enterprise-grade data persistence, user profile history tracking, and direct integration with major hospital electronic health record (EHR) systems to help patients track their health metrics over long periods.
Log in or sign up for Devpost to join the conversation.