作者: Vinh-Thong Ta , Rémi Giraud , D. Louis Collins , Pierrick Coupé
DOI: 10.1007/978-3-319-10443-0_14
关键词:
摘要: Automatic segmentation methods are important tools for quantitative analysis of magnetic resonance images. Recently, patch- based label fusion approaches demonstrated state-of-the-art accuracy. In this paper, we introduce a new patch-based method using the PatchMatch algorithm to perform anatomical structures. Based on an Optimized PAtchMatch Label (OPAL) strategy, proposed provides competitive accuracy in near real time. During our validation hippocampus 80 healthy subjects, OPAL was compared several methods. Results show that obtained highest median Dice coefficient (89.3%) less than 1 sec per subject. These results highlight excellent performance terms computation time and recently published