Local Similarity Enhancement For The Computational Segmentation Of The Right Atrium In Cardiac Computed Tomography

This work proposes a strategy to segment the right atrium (RA) into three-dimensional (3-D) multi-layer Computed Tomography (CT) images. This strategy consists of the stages of pre-processing, segmentation and intonation of parameters. The pre-processing stage is divided into two phases. In the first one, called the filtering phase, a technique called Local Similarity Enhancement (LSE) is used in order to reduce the impact of 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). In the second, identified as the phase of definition of a Region Of Interest (ROI), we consider filtered images, least squares vector support machines and a priori information to isolate the anatomical structures that surround the RA. On the other hand, a clustering algorithm, called Region Growth (RG), is implemented during the 3-D segmentation stage, which is applied to the preprocessed images. During the intonation of parameters of the proposed strategy, the Dice coefficient (Dc) is used to compare the segmentations, obtained automatically, with the segmentation of the right atrium generated manually by a cardiologist. The combination of parameters that generated the highest Dc considering the instant of diastole is then applied to the remaining 19 three-dimensional images, obtaining an average Dc higher than 0.85 which indicates a good correlation between the segmentations generated by the expert cardiologist and those produced by the strategy developed.