Reproduce “Pathology Randomization for Synthetic Priors and Zero-Shot Detection” 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
Alexandra G. Roberts, Tonmoy Hossain, Mark Olchanyi, Stefan Brandstätter, Jonathan Williams Ramirez, Ana Lawry Aguila, Matthew Rosen, Annabel Sorby-Adams and 2 more
PathoDraw trains a contrast-agnostic brain-MRI detector using synthetic lesions drawn into healthy anatomical labels and paired healthy and altered renderings. The network predicts underlying healthy anatomy and signed image deviations, whose magnitude flags departures from a learned healthy prior. On BraTS T1-weighted scans, its ceiling Dice is 0.537 versus FAE’s 0.437, an oracle-threshold result, not a clinical operating point or evidence of universal pathology coverage.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.