Published March 19, 2020 | Version v2
Other Open

Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge

  • 1. Universitat de Barcelona, Spain
  • 2. Vall d'Hebron Hospital, Barcelona, Spain
  • 3. Clinical University Hamburg, Germany
  • 4. McGill University Health Centre, Montreal, Canada


This is the challenge design document for the "Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge", accepted for MICCAI 2020.

In the recent years, many machine/deep learning models have been proposed to accurately segment cardiac structures in magnetic resonance imaging [1-4]. However, when these models are tested on unseen datasets acquired from distinct MRI scanners or clinical centres the segmentation accuracy can be greatly reduced [5,6]. This makes it difficult for these tools to be applied consistently across multiple clinical centres, especially when subjects are scanned using different MRI protocols or machines. The M&MS challenge is the first international competition to date on cardiac image segmentation combining data from different centres, vendors, diseases and countries at the same time. It will evaluate the generalisation ability of machine/deep learning and cross-domain transfer learning techniques for cardiac image segmentation, by testing these on a cohort of 350 cardiac MRI studies comprising healthy, hypertrophic and dilated hearts, acquired in three different countries (Spain, Germany and Canada) and by using four distinct MRI vendors (Siemens, Philipps, General Electric and Canon). The challenge will be supported by the H2020 euCanSHare project (, which is building a multi-centre big data platform for cardiovascular personalised medicine research. The three top teams will receive prizes of 500, 300 and 200 Euros, respectively.


[1] Bernard, Olivier, et al. "Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: Is the problem solved?." IEEE transactions on medical imaging 37.11 (2018): 2514-2525.
[2] Tran, Phi Vu. "A fully convolutional neural network for cardiac segmentation in short-axis MRI." arXiv preprint arXiv:1604.00494 (2016).
[3] Isensee, Fabian, et al. "Automatic cardiac disease assessment on cine-MRI via time-series segmentation and domain specific features." International workshop on statistical atlases and computational models of the heart. Springer, Cham, 2017.
[4] Zotti, Clément, et al. "GridNet with automatic shape prior registration for automatic MRI cardiac segmentation." International Workshop on Statistical Atlases and Computational Models of the Heart. Springer, Cham, 2017.
[5] Tao, Qian, et al. "Deep learning–based method for fully automatic quantification of left ventricle function from Page 1 of 12Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge cine MR images: a multivendor, multicenter study." Radiology 290.1 (2018): 81-88.
[6] Dangi, Shusil, Ziv Yaniv, and Cristian Linte. "A Distance Map Regularized CNN for Cardiac Cine MR Image Segmentation." arXiv preprint arXiv:1901.01238 (2019).
[7] Zhuang, Xiahai et al. “Evaluation of algorithms for Multi-Modality Whole Heart Segmentation: An open-access grand challenge.” Medical Image Analysis (2019).