Segmentación automática de la arteria aorta ascendente y la válvula aórtica en imágenes de tomografía computarizada cardiaca/Automatic segmentation of the ascending aorta and aortic valve in computed tomography images

• Published in: Revista latinoamericana de hipertensión Vol. 12; no. 2; p. 70
• Main Authors: Vera, Miguel, Huérfano, Yoleidy, Contreras-Velásquez, Julio, Del Mar, Atilio, Rodríguez, Johel, Bautista, Nahid, Wilches-Durán, Sandra, Graterol-Rivas, Modesto, Riaño-Wilches, Daniela, Rojas, Joselyn, Bermúdez, Valmore
• Format: Journal Article
• Language: English
• Published: Caracas Sociedad Latinoamericana de Hipertension 01.04.2017
• Subjects: Cardiology; Coronary vessels; Heart; Medical imaging; Three dimensional imaging; Tomography
• ISSN: 1856-4550

The present work proposes a technique for the automatic segmentation of the anatomic set consisting of the ascending aorta + aortic valve (AAAV) in 10 three-dimensional (3-D) cardiac images of multi-cut computed tomography, belonging to the same subject. The mentioned technique consists of the stages of pre-processing and segmentation. The pre-processing stage includes two phases: the first, minimizes both Poisson noise and the impact of the staircase artifact, we use a technique called global similarity enhancement, this type of enhancement consists of the application of a bank of filters, softeners And a border detector, whose purpose is to generate an image in which the information of the anatomical structures, which make up the original images, is grouped together; the second phase, considering the filtered images, uses a priori information about the location of the aortic valve and a learning paradigm, based on vector support machines, to define a region of interest that isolates AAAV from neighboring anatomical structures. On the other hand, to generate the 3-D morphology of the TAA, a segmentation stage is applied which considers the filtered images and a clustering algorithm based on regions growth. The proposed strategy generates the 3-D segmentations of AAAV in all the images that make up the complete cardiac cycle of the subject considered. In order to quantify the performance of the referred technique, the Dice coefficient was considered, obtaining a good correlation between the automatic segmentations and the manual ones generated by a cardiologist. Automatically generated segmentations may be helpful in detecting certain pathologies that affect both the aorta and anatomical structures associated with it, such as the aorta and leftventricle.