Reproduce “RT-Super: Learning Tumor Segmentation from Longitudinal Images and Reports” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Pedro R. A. S. Bassi, Wenxuan Li, Hanxue Gu, Jieneng Chen, Xinze Zhou, Zheren Zhu, Sezgin Er, Ibrahim E. Hamamci and 6 more
RT-Super uses reports and multiple CT scans to train a teacher that distills segmentation supervision to a student using one scan at test time. The authors evaluate it on spleen, esophagus, and uterus tumors, including an unseen hospital. Its report-only variant achieves mean external DSC of 30, versus 16 for report-only R-Super, on a separate annotated set.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.