Assessment of particular abdominal aorta section extraction from contrast-enhanced computed tomography angiography

The aim of this work is to improve the accuracy of extraction of a particular abdominal aorta section and to reduce the distortion in three-dimensional Computed Tomography Angiography (CTA) images. Imaging modality and quality plays crucial role in the medical diagnostic process, thus ensuring high quality of images is essential at every stage of acquisition and processing.Noise is defined as a disturbance of the image quality that characterized by the presence of pixels that are not desired in the image. Removing or reducing the occurrence of the noise is still a challenging problem in the image processing field. We selected denoising or smoothing methods, such as Gaussian, anisotropic, Taubin, and Laplace filter. This work was carried out as a part of the aorta segmentation software which can detect and display abdominal aorta ROI. We embedded the denoising methods in the system in order to smoothen and to enhance the detection of the contour and surface of the vasculature, which subsequently leads to more accurate segmentation results and extraction particular segment of aorta. We compare the performance of method by using the aortic segmentation result represented in the form of the Dice Similarity Coefficient. Gaussian smoothing is one noise reduction technique that obtained the best in segmentation (average Dice similarity coefficient 0.9103 ±0.04. The results showed that the segmentation results presented in this work are promising, although not all the visualization was well maintained.