Regresar

Enhancement by local similarity for automatic segmentation of the thoracic aorta in cardiac computed tomography

Abstract:

This work proposes a strategy to segment the thoracic aortic artery (TAA) into three-dimensional (3-D) multi-layer computed tomography images. This strategy consists of the stages of filtering, segmentation and intonation of parameters. The filtering stage employs a technique called local similarity enhancement (LSE) in order to reduce the impact of the artifacts and attenuate noise in the quality of the images. This technique combines an averaging filter, an edge detector filter (called black top hat) and a Gaussian filter (GF). On the other hand, a clustering algorithm, called region growth (RG), is implemented during the 3-D segmentation stage, which is applied to the pre-processed images. During the intonation of parameters of the proposed strategy, the Dice coefficient (Dc) is used to compare the segmentations, of the TAA, obtained automatically, with the segmentation of the TAA generated, manually, by a cardiologist. The combination of parameters that generated the highest Dc considering the instant of diastole is then applied to the 9 remaining three-dimensional images, obtaining an average Dc higher than 0.92 which indicates a good correlation between the segmentations generated by the expert cardiologist and those produced by The strategy developed.