Reproduce “Training-Time Domain Transfer for Generator-Free Diagnostics on Novel Modalities in Computational Pathology” 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
Julia A. Mielcarz, Andrea Camilloni, Nita Mulliqi, Jessica A. Frank, Anders Blilie, Emilius A.M. Janssen, Martin Eklund, Kimmo Kartasalo
The authors introduce training-time domain transfer (TTDT) for prostate-biopsy diagnosis on unstained tissue. A Pix2Pix generator translates paired H&E slides into synthetic raw or deparaffinized training examples; a diagnostic model then analyzes real unstained images, without generation at inference. In a 79-slide, single-laboratory test, TTDT agreed more with the H&E model than virtual staining, without establishing clinical equivalence.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.