HVSMR-2.0 Dataset (released 2024)

HVSMR-2.0 is a dataset for whole-heart segmentation from 3D cardiovascular MR images in patients with congenital heart disease. The dataset is an extension of our HVSMR challenge (see below).

HVSMR-2.0 contains 60 cardiovascular MR images with even more heart defects, and separate segmentations of eight structures: the four cardiac chambers and four great vessels.

The dataset is described in Nature Scientific Data (see [PDF]) and available on figshare.

MICCAI 2016 HVSMR Challenge

The challenge on Whole-Heart and Great Vessel Segmentation from 3D Cardiovascular MRI in Congenital Heart Disease (HVSMR) was held at MICCAI 2016 and co-organized by myself, Mehdi Hedjazi Moghari, Alireza Akhondi-Asl and Andrew Powell.

The dataset contained 20 cardiovasular MR images from patients with a variety of congenital heart defects, and segmentations of two structures: the (global) blood pool and the ventricular myocardium. The submission system and leaderboards are no longer active.