OncoBoard.ai

OncoBoard.ai is a multi-agent AI system built with Google Gemini that transforms how cancer centers prepare for multidisciplinary tumor boards. What currently takes a nurse coordinator 8–12 hours of manual work—pulling patient records, reviewing imaging, checking clinical guidelines, searching trials, and writing summaries—now happens in under 8 minutes through seven specialized AI agents working in parallel. The system operates in three phases with human oversight at each gate: pre-meeting agents compile comprehensive case data from radiology, pathology, genomics, treatment guidelines (NCCN/ESMO), and active clinical trials; during-meeting agents provide live transcription and capture consensus recommendations; post-meeting agents draft clinical notes, dispatch action items, and track follow-up completion. A RAG-powered clinical chat allows clinicians to query patient data in real-time during the board. Built on FastAPI + Vue.js and deployed on Vercel with 1,097 TCGA-BRCA cases, OncoBoard.ai keeps doctors focused on what only they can do: making the clinical call.