Pan-cancer CT segmentation dataset adds 17,000 labeled cases and a new challenge
yuyinzhou_cs · x · 2026-07-23
Largest pan-cancer CT segmentation dataset released
The team released a pan-cancer CT segmentation dataset with 17,000+ labeled cases and launched a Foundation Models for Pan-Cancer Segmentation in CT Images Challenge.
- The dataset aggregates publicly available cancer CT datasets with lesion annotations.
- It is meant to support more scalable and generalizable AI models for cancer imaging.
- Links were shared for both the dataset and challenge submission.
More from Research
- AnyTop can generate motion for unseen skeletons with just three examples per topology — rsasaki0109 · 2026-07-23
- Garfield corpus analysis tracks 3,654 Codex files and 1,499 transcripts — zeeg · 2026-07-23
- A new multi-view 3D head reconstruction paper uses coarse-guided surface sampling — maier_ak · 2026-07-23
- A new blog post argues attribution should follow the geometry used to train the model — aryaman2020 · 2026-07-23
- U.S. Energy Department picks first Genesis Mission projects for AI science — donutloop · 2026-07-23
- Self Gradient Forcing adds missing memory-writing gradients for long video extrapolation — Junhao Zhuang · 2026-07-23