Delineation of LV using Local Motion Intensity Clustering Technique

Abstract

Magnetic Resonance Imaging (MRI) is a powerful tool for identification of finding anatomical information and functionality of heart. The delineation of the LV in MRI images is a promising task because of repetitive and fast flow of blood in the heart. For such images noise interference is challenging problem. To Process such a challenge, an algorithm for delineating LV is proposed here. The proposed approach LMIC model connects the Local Intensity Clustering with mobility intensity information of LV. The Lucas Kanade optical flow way is used to calculate motion intensity of LV. This motion intensity is used as energy parameter for clustering purpose and further by using level set method accurate separation of LV has done.  The LV can be accurately segmented by this approach. The experimentation has been using parameters such as dice metric and Haussdorff distance.  Here Experimental result shows better model to delineate LV over existing methods to address the problem of image noise nosiness.