Registration-based automatic 3D segmentation of cardiac CT images

Abstract

Segmenting whole heart from cardiac computed tomography (CT) images can provide basis for evaluation of cardiac function and help improve accuracy of clinical diagnosis. Manual segmentation is the most accurate method for cardiac segmentation. But it is time consuming and not sufficiently reproducible. However, clinicians still rely on this method in practical applications. So a fully automatic method is needed to improve the segmentation efficiency. This paper proposes a registration-based automatic approach for three-dimensional (3D) segmentation of cardiac CT images. The proposed method utilizes the similarity of cardiac CT images between different individuals, and uses registration to achieve segmentation. Affine transformation is first implemented to achieve global coarse registration. Then, cubic B-splines are used to refine the local details in locally accurate registration. Mutual information (MI) is used as the similarity measure, and adaptive stochastic gradient descent (ASGD) as the optimization algorithm. Our method is applied to the dual-source cardiac CT images to segment whole heart. Experimental results show that the proposed method can automatically segment whole heart from cardiac CT images. © 2013 Springer-Verlag.