Open-source cancer model from JHU and Stanford hits every tumor across 34 public CTs
MaziyarPanahi · x · 2026-11-20
The author ran Merlin-Cancer-Super, developed by Johns Hopkins and Stanford, on 34 public CTs of women with uterine cancer with no hints, on a single GPU. It touched every tumor; one segmentation landed inside the radiologists' outline with Dice 0.83. On 30 of 34 cases it drew the tumors smaller than the radiologists did.
More from Models
- Open-source cancer model hits tumors in 34/34 CTs but median Dice is only 0.22 — MaziyarPanahi · 2026-11-20
- Gemini 4 Neon flunks Runescape Bench: ambitious strategies, terrible time management — morgymcg · 2026-11-20
- Fan makes an AI video showing how much ~30B Qwen models improved in 28 months — vanstriendaniel · 2026-11-20
- moyix on LLMs' worst writing habit: "disclaiming things no one ever claimed" — moyix · 2026-11-20
- Users Praise 6.2 Astra as Tender, Introspective, and Shockingly Perceptive — repligate · 2026-11-20
- Mystery model Gestalt isn't Fable 5.5 and belongs to no major AI lab, rumor thread claims — garrytan · 2026-11-20