Other Open Access
Xiahai Zhuang; Lei Li
This is the challenge design document for the "Multi-sequence CMR based myocardial pathology segmentation challenge", accepted for MICCAI 2020.
Assessment of myocardial viability is essential in the diagnosis and treatment management for patients suffering from myocardial infarction (MI). Cardiac magnetic resonance (CMR) is particularly used to provide imaging anatomical and functional information of heart, such as the late gadolinium enhancement (LGE) CMR sequence which visualizes MI, the T2-weighted CMR which images the acute injury and ischemic regions, and the balanced-Steady State Free Precession (bSSFP) cine sequence which captures cardiac motions and presents clear boundaries. Combining these multi-sequence CMR data can provide rich and reliable information as well as morphological information of the myocardium.
The target of this challenge is combining multi-sequence CMR data to segment each position of the myocardium into different pathologies, including normal myocardium, infarction and edema. This is referred to as the myocardium pathology classification/segmentation, which is crucial for the diagnosis and treatment management of patients. This is, however, still arduous, particularly due to the pathological myocardium. Since manual delineation is generally time-consuming, tedious and subjects to inter- and intra-observer variations, automating this segmentation is highly desired in clinical practice.
The proposed challenge will provide the three sequence CMR images from 45 patients [1, 2], for developing novel algorithms that can segment myocardial pathology combining the complementary information from these threesequence CMR images. The challenge presents an open and fair platform for various research groups to test and validate their methods on these datasets acquired from the clinical environment. The aim of the challenge will not only be to benchmark various myocardial pathology segmentation algorithms, but also to cover the topic of general cardiac image segmentation, registration and modeling, and raise discussions for further technical development and clinical deployment.
1] Xiahai Zhuang. "Multivariate mixture model for myocardial segmentation combining multi-source images." IEEE transactions on pattern analysis and machine intelligence 41.12 (2018): 2933-2946.
 Xiahai Zhuang. "Multivariate mixture model for cardiac segmentation from multi-sequence MRI." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2016.
 Lei Li, et al. "Atrial scar quantification via multi-scale CNN in the graph-cuts framework." Medical Image Analysis 60 (2020): 101595.