||In athletes such as cricket fast bowlers and javelin throwers, asymmetries of trunk muscles can be observed and measured using MR imaging. These asymmetries are of interest in sports medicine and science because they may be predictive of future sports injury. Measurement of these asymmetries requires that the muscles be segmented in the MR images. This is a challenging task, because of the poor contrast between adjacent muscles.
In this talk, I will present a semi-automatic segmentation approach for this purpose. In a recent study, we demonstrated that the segmentations obtained with this approach are qualitatively similar to manual segmentations, while intra-user variations are reduced by more than 50%.