Automatic Segmentation of Coronary Arteries Based on Region Growing and Discrete Wavelet Transformation

Abstract

In this work, we proposed a new method for segmenting coronary arteries automatically in computed tomography angiography (CTA) data sets which allows the coronary pathologies to be reliably detected. Region growing is firstly applied to give the initial segmentation of coronary arteries. However, most applications in the field of medical image processing require precise estimation. Based on the initial segmentation obtained from region growing, discrete wavelet transformation (DWT) completes accurate detection of coronary arties. The proposed method is tested on CTA data-sets. The experimental results indicate that the proposed method has good performance.