Inspiration
What it does
it uses searching tools to search revitalize diseases to what you have
How we built it
🔬 Reka's multimodal vision model powers our medical image analysis agent — it reads X-rays, CT scans, and MRIs, extracting structured findings with region-level confidence scores directly from pixel data. 🔎 Tavily's real-time search API drives both our Research Agent and Epidemic Surveillance Agent — it surfaces the latest medical literature, outbreak news, and disease wave data so diagnoses are grounded in what's happening right now, not just static knowledge. 🧠 Claude (Anthropic) acts as the Coordinator, synthesizing findings from all agents into a unified diagnostic report with differential diagnoses and confidence rankings. 📚 Yutori Research & Browsing APIs complement Tavily by performing deep-dive PubMed searches and browsing medical portals for peer-reviewed citations and raw research data. 🏥 Project Consilium is a multi-agent diagnostic pipeline: patient intake → concurrent AI agents (Vision via Reka, Research via Tavily/Yutori, Epidemic Surveillance via Tavily/disease.sh, Knowledge Graph) → Claude synthesis → comprehensive clinical report. ⚡ Built with Next.js 16, React 19, FastAPI, Tailwind CSS 4, and shadcn/ui. Real-time pipeline tracking shows users exactly what each agent is doing as it runs.
Log in or sign up for Devpost to join the conversation.