Speech title：Coronary Artery Segmentation in Medical Image

Abstract: We proposed a new segmentation method to extract coronary vessels. We investigate the efficiency of U-net, which has been shown to be   very successful in medical image segmentation, for coronary vessel segmentation. Specially, we propose for iterative segmentation based on U-Net to extract the coronary vessels in various sizes. We use U-Net to segment the vessels. Based on information on the extracted vessels, we introduce a new approach to extract junctions on the vascular tree and small vessel regions. The proposed approach is evaluated using a coronary X-ray angiography dataset. The experimental results show that our approach can outperform the standard method and attain the accuracy of 76.40%.

Biography

1995: B.S., 2000: M.S., 2009: Ph.D. degree in Computer Science from University of Science, Vietnam

1995 ~ 2018: Lecturer and Professor in Department of Computer Science, Faculty of Mathematics & Computer Science, University of Science, Vietnam

2019~now: Dean and Professor in Department of Computer Science, Information Science Faculty, Sai Gon University.