Reproduce “SegDT: A Diffusion Transformer-Based Segmentation Model for Medical Imaging” 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
Salah Eddine Bekhouche, Gaby Maroun, Fadi Dornaika, Abdenour Hadid
SegDT segments skin lesions in dermoscopic images using an extra-small diffusion transformer and a velocity-based reverse process called rectified flow. On ISIC 2016, it reports 94.76% Dice, versus 92.46% for DU-Net+. The authors report comparable quality to IDDPM in 15 rather than 35 steps, but timing conditions are unspecified; results concern benchmarks, not clinical benefit.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.