Heart vessel extraction using motion based heart area extraction

Abstract

In this paper, a new method for heart vessel extraction based on heart area segmentation in angiogram image sequences is presented. One of the difficulties in vessel extraction in angiogram images is the detection of ribs and spins as vessels. Therefore, to cope with this problem, we utilize motion information to segment heart area. Then a two-stage vessel extraction algorithm is utilized to extract heart vessels efficiently. The vessel extraction algorithm, firstly extracts thick vessel, then the vessel extraction is completed by utilizing a new morphology based algorithm. Experimental results show that the proposed algorithm extracts heart area precisely using motion based segmentation, and hence vessel extraction accuracy of 95.16% is obtained.