Reproduce “Registration is a Representation Problem: Solving Whole-Body PET/CT Registration” 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
Neel Dey, Adrian V. Dalca
The method registers longitudinal whole-body PSMA PET/CT scans while limiting warp-induced tumor-volume change. It combines learned anatomix features, handcrafted MIND-SSC descriptors, organ and lesion labels, and lesion-aware volume constraints in a two-stage optimization pipeline, without task-specific model training. On the Learn2Reg 2026 PSMAReg leaderboard, it ranked first with a composite score of 71.4, but this result is limited to that challenge.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.