Archives

REG2025 Debug Phase dataset

This project advances automated pathology report generation using vision-language foundation models. It goes beyond traditional NLP metrics by focusing on clinically meaningful evaluation. The challenge uses a diverse dataset of ~20,500 cases from six countries and promotes integration into clinical workflows with expert feedback.

BFA Phase 1: Trial Images dataset

COSAS Docker Testing Phase dataset

Demo archive

Demo archive with radiology images

STOIC2021 public training image

One CT scan from the public STOIC2021 training set: https://registry.opendata.aws/stoic2021-training/

KiTS21 Sanity Check

The three "sanity check" cases for the KiTS21 submission period

LUNA16

888 CT scans from the LUNA16 challenge

LOLA11

55 scans from the LOLA11 challenge

coronacases.org

10 CT scans from the website https://coronacases.org/

RadboudCOVID

Data from RadboudUMC from Covid-19 (suspected) subjects