Segmentation of cardiac left ventricle from MRI images using a method based on FCM and convexity preserving level set

摘要

Segmentation of the left ventricle (LV) is of great significance in diagnosis and treatment of cardiac diseases. In this paper, a rapid method is proposed for segmentation of left ventricle from MRI images. Specifically, each MRI image is divided into two clusters, the background and the foreground, using a pixel-based fuzzy c-mean (FCM) method firstly. Then each connected region is assigned with different connected component label (CCL) so that the label most similar to the left ventricle is selected based on the multiple attributes such as center distance, eccentricity and area size. Finally, a convexity preserving level set method is used to optimize and extract the left ventricular region. Experiment results show that the proposed method is accurate and effective in segmentation of the left ventricle with a fast average computational time of 1.02 seconds per image, which is superior to some current advanced methods.