A 3D anatomy atlas driven by plain language. Ask about cerebral infarction and the AI picks the relevant organs and vessels, renders them in 3D and adds an explanation. Interpretation, identification and drawing are separate stages, holding down wrong displays, and ambiguous input comes back as candidates to confirm.
Judges praised learning through dialogue while organs are shown in 3D, the beauty of the interface, and the care taken to lower cognitive load. They saw a new route opening in a field that has relied on physical models, with classroom use in view.
In medical study the worst failure mode is a diagram that looks plausible and is wrong. Separating interpretation, identification and rendering, and returning candidates instead of guessing at ambiguous input, reads as an attempt to prevent that failure structurally. For a grand prize, presentation alone rarely suffices; what likely separated this entry is that beneath an appealing surface sits a foundation that does not break when it is put in front of learners.
See all winners of 第4回 Agentic AI Hackathon with Google Cloud →
Summaries are written by this site. Project pages include demo videos, screenshots and the team's own write-up (videos may autoplay).