Title :
Accurate segmentation for right ventricles based on Multi-Atlas
Author :
Yaonan Zhang ; Xiangfei Meng ; Chuanshen Chen ; Qian Song ; Yu Xu
Author_Institution :
Sino-Dutch Biomed. & Inf. Eng. Sch., Northeastern Univ., Shenyang, China
Abstract :
A new image registration and fusion method based on Multi-Atlas was proposed in this thesis. It treats all Atlas images as a series of data nodes and clusters them through massage propagation to obtain different exemplars, then uses these exemplars to participate in the normal atlas segmentation, orders them based on the dice similar coefficient, iterate, gets the final accurate segmentation. In term of registration, three-step registration is adopted: first rigid registration, then affine registration and B spline registration. Besides, fusion strategy called Joint Label Fusion is used, which is first used for right ventricle. The method makes split operation by registration and fusion method based on Multi-Atlas image, corrects a Multi-Atlas label fusion error caused by the registration, thereby optimizes segmentation result.
Keywords :
biomedical MRI; cardiology; image fusion; image registration; image segmentation; medical image processing; Atlas images; B spline registration; Multi-Atlas label fusion error; affine registration; data clusters; data nodes; dice similar coefficient; image fusion method; image registration method; joint label fusion strategy; normal atlas segmentation; right ventricle segmentation; rigid registration; segmentation optimization; three step registration; Biomedical imaging; Feature extraction; Heart; Image registration; Image segmentation; Splines (mathematics); Transforms; Atlas selection; Multi-Atlas; image registration; label fusion;
Conference_Titel :
Medical Imaging Physics and Engineering (ICMIPE), 2013 IEEE International Conference on
Conference_Location :
Shenyang
Print_ISBN :
978-1-4799-6305-8
DOI :
10.1109/ICMIPE.2013.6864527