Reproduce “TME-ContextNet: Class-Conditioned Spatial Supervision and Grouped Decoding for Joint Nuclei Instance Segmentation and Classification” 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
Rukang Ma, Bingxue Wang, Liwen Zou, Yingying Cao, Zhenghua Cai, Yu Li, Dong Wang, Huiyu Zhou and 1 more
TME-ContextNet jointly detects and types nuclei in H&E pathology images. It derives class-conditioned spatial supervision from existing nuclear labels, then uses a grouped decoder that shares features before separating geometry and semantic tasks; the auxiliary head is removed at inference. On PanNuke, it reports mPQ+ of 43.74 ± 2.89%, versus CellViT’s 41.91 ± 0.30%, a 1.83-point margin; results vary by dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.