Reproduce “Improving PI-RADS Prediction and Prostate Cancer Detection from Multiparametric MRI with Anatomical Guidance” 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

Seungbin Park, Thibault Heintz, Rachit Saluja, Francesca Khani, Timothy McClure, Daniel Margolis, Mert R. Sabuncu, Heejong Kim

The study tests whether adding prostate-zone masks to MRI inputs improves patient-level PI-RADS prediction and clinically significant prostate cancer detection. Models use two- or five-zone masks. On testing, ResNet-18 with two-zone masks improved PI-RADS agreement, while ResNet-50 with five-zone masks improved cancer discrimination. Combining a model-predicted PI-RADS score with an expert score improved discrimination; external tests were fine-tuned and covered scores 3–5.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.