ABSTRACT
Accurate and fast whole cardiac substructures segmentation from Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) is crucial in developing clinical applications, such as computer-aided surgery and computer-aided diagnosis. However, the segmentation of different substructures is challenging because of the amount of data that should be annotated by experts, the diversity of sizes and shapes of cardiac substructures, and the complexity of the background. This work aims to develop an automatic and fast whole heart segmentation including all cardiac substructures as well as the great vessels. The proposed approach used mainly image processing methods that enable the heart segmentation from sagittal, axial, and coronal views to obtain a 3D reconstruction. Finally, the experiments are conducted on both Automated Cardiac Diagnosis Challenge and CT scans acquired from a patient with COVID-19 at the Cheikh Zaid International University Hospital in Rabat Morocco. © 2022 IEEE.