报告时间： 2019年12月13日9:30-11:30 ，星期五(Friday)
(School of Computer Science Building, Room 4001)
邀 请 人： 肖亮 (Liang Xiao)
9:30am - 10:30 am
Report: Prof. Mohamed Abdel-Mottaleb，Digital Image Processing in Ophthalmology
10:30 am - 11:00 am
Report：Prof. Liang Xiao, A brief introduction of Jiangsu Key Lab of Spectral Imaging and Intelligent Sensing
11:00 am -12:00 am
Demo in the lab
Prof. University of Miami ,IEEE Biometrics Compendium的主编，IEEE Fellow
Digital Image Processing in Ophthalmology
In this presentation, we will describe our work on using digital image processing in ophthalmology applications. The first application deals with detecting glaucoma from fundus images of the eye. We developed several deep learning models using transfer learning, semi supervised and self-learning. We show that using deep learning and only fundus images, we obtain performance higher than expert ophthalmologists by a large margin. The second application deals with the segmentation of the eye’s corneal layers and measuring their thickness from Optical Coherence Tomography (OCT) images. This is important because various corneal diseases, such as Fuchs dystrophy, Keratoconus and corneal graft rejection, can be diagnosed based on the changes of the thickness of corneal microlayers. OCT technology made it possible to obtain high resolution corneal images that show the microlayer structures of the cornea. Manual segmentation is subjective and not practical due to the large volume of obtained images. Existing automatic methods, used for segmenting corneal layer interfaces, are not robust and segment fewer corneal microlayer interfaces compared to our method. Our approach segments all the layers of the cornea and the experimental results show its robustness. In the third application we present a system using VR/AR for measuring the visual field of patients with glaucoma and macular degeneration. The system is also used for aiding those patients by manipulating scene images in order to increase their visibility.
Bio：Mohamed Abdel-Mottaleb received the Ph.D. degree in computer science from the University of Maryland, College Park, in 1993. He joined the University of Miami in 2001. Currently, he is a Professor and Chairman of the Department of Electrical and Computer Engineering. His research focuses on biometrics, visual tracking, human activity recognition, and medical image processing. Prior to joining the University of Miami from 1993 to 2000, he was with Philips Research, Briarcliff Manor, NY, where he was a Principal Member of the Research Staff and a Project Leader. At Philips Research, he led several projects in image processing and content-based multimedia retrieval. He represented Philips in the standardization activity of ISO for MPEG-7, where some of his work was included in the standard. He holds 22 U.S. patents and more than 30 international patents. He published more than 150 journal and conference papers in the areas of image processing, computer vision, and content-based retrieval. He serves as the Editor in Chief for the IEEE Biometrics Compendium. He is an IEEE fellow since January 2011.
BIO：Mohamed Abdel-Mottaleb于1993年获得马里兰大学公园分校的计算机科学博士学位。他于2001年进入迈阿密大学。目前，他是电气和计算机工程系的教授兼主任。他的研究重点是生物识别，视觉跟踪，人类活动识别和医学图像处理。在1993年至2000年进入迈阿密大学之前，他在纽约州Briarcliff Manor的飞利浦研究中心是研究人员的首席成员和项目负责人。在飞利浦研究中心，他领导了多个图像处理和基于内容的多媒体检索项目。他代表飞利浦参加了MPEG-7的ISO标准化活动。他拥有22项美国专利和30多项国际专利。他在图像处理，计算机视觉和基于内容的检索领域发表了150多篇期刊和会议论文。他是IEEE Biometrics Compendium的主编。自2011年1月起担任IEEE研究员。