Semi-automatic segmentation method of left ventricle in cardiac cine MRI for analysis of myocardial viability

Cardiovascular disease is one of the major health problems in the world. The complexity of cardiac cine MR images and on the other hand, the diversity of cardiovascular disease require segmentation phase. This phase is an important step in the processing and interpretation of medical images. In this paper, a new approach for the segmentation of the left ventricle from short-axis cardiac cine MR images is presented. Our approach is based on three-steps: In the first one, we have effected a pretreatment using mathematical morphology to improve images quality. Thereafter, we proceed on the segmentation of the left ventricle, in this phase we combined between the region growing and the active contour methods to obtain the endocardial form exactly. Finally, the parameters were calculated for characterization, such as tele-diastolic volume (VTD), tele-systolic volume (VTS) and ejection fraction (FE) to predict cardiovascular diseases. The proposed method was tested on the Heart-database containing 18 patients. The results are satisfactory with excellent endocardial segmentation and a good correlation of 88% for VTD and 91% for VTS between our parameters and those of experts which shows the good performance of our approach which can be used to aid diagnosis.