报告时间：星期五（2018.01.19）15:00
报告地点：电子信息楼306
报告人：Y Jiang Li, Associate Professor, Old Dominion University
邀请人：陈新建特聘教授
报告内容简介：Post-traumatic stressdisorder (PTSD) is a traumatic-stressor related disorder developed by exposureto a traumatic or adverse environmental event that caused serious harm orinjury. Structured interview is the only widely accepted clinical practice forPTSD diagnosis but suffers from several limitations including the stigmaassociated with the disease. Diagnosis of PTSD patients by analyzing speechsignals has been investigated as an alternative since recent years, wherespeech signals are processed to extract frequency features and these featuresare then fed into a classification model for PTSD diagnosis. In this paper, wedeveloped a deep belief network (DBN) model combined with a transfer learning(TL) strategy for PTSD diagnosis. We computed three categories of speechfeatures and utilized the DBN model to fuse these features. The TL strategy wasutilized to transfer knowledge learned from a large speech recognitiondatabase, TIMIT, for PTSD detection where PTSD patient data is difficult tocollect. We evaluated the proposed methods on two PTSD speech databases, eachof which consists of audio recordings from 26 patients. We compared theproposed methods with other popular methods and showed that thestate-of-the-art support vector machine (SVM) classifier only achieved anaccuracy of 57.68%, and TL strategy boosted the performance of the DBN from61.53% to 74.99%. Altogether, our method provides a pragmatic and promisingtool for PTSD diagnosis.
报告人简介：Dr. Jiang Li is an Associate Professor in theDepartment of Electrical and Computer Engineering at Old Dominion University(ODU). He received his Ph.D. degree in electrical engineering from theUniversity of Texas at Arlington, TX, in 2004, the M.S. degree in automationfrom Tsinghua University, China, in 2000, and the B.S. degree in electricalengineering from Shanghai Jiaotong University, China, in 1992. His researchinterests include pattern recognition, computer-aided medical diagnosis systems,image processing and deep learning. Dr. Li has published more than 100 journaland conference papers. He is a reviewer for several journals and a guestassociate editor for Medical Physics. Dr. Li worked as a postdoctoral fellow at the department of radiology,National Institutes of Health, from 2004 to 2006. He joined ODU as an assistantprofessor in Spring 2007. He is a senior member of the IEEE.