題目：Big Data Analytics for Health Informatics: Detecting Adverse Drug Events from Electronic Health Records
主講人：魏志平教授（臺灣大學）
時間：2015.5.29（周五）上午9:00-11:00
地點：主樓418
主講人簡介：
Chih-Ping Wei received a BS in Management Science from National Chiao-Tung University in Taiwan, R.O.C. in 1987 and an MS and a Ph.D. in Management Information Systems from the University of Arizona in 1991 and 1996, respectively. He is currently a distinguished professor of Department of Information Management at National Taiwan University, Taiwan, R.O.C. Prior to joining National Taiwan University in 2010, he was a professor of Institute of Service Science and Institute of Technology Management at National Tsing Hua University in Taiwan and a professor of Department of Information Management at National Sun Yat-sen University in Taiwan. He was also a visiting scholar at the University of Washington in Fall 2013, the University of Illinois at Urbana-Champaign in Fall 2001, and Chinese University of Hong Kong in Summer 2006 and 2007. His papers have appeared in Journal of Management Information Systems (JMIS), European Journal of Information Systems, Decision Sciences, Decision Support Systems (DSS), IEEE Transactions on Engineering Management, IEEE Software, IEEE Intelligent Systems, IEEE Transactions on Systems, Man, and Cybernetics, IEEE Transactions on Information Technology in Biomedicine, Journal of the American Society for Information Science and Technology, Information Processing and Management, Journal of Database Management, and Journal of Organizational Computing and Electronic Commerce, etc. His current research interests include data analytics and business intelligence, text mining and information retrieval, patent analysis and mining, and health informatics. He has edited special issues for Decision Support Systems, International Journal of Electronic Commerce, Electronic Commerce Research and Applications, Information Processing and Management, Pacific Asia Journal of the Association for Information Systems, and Information Systems and e-Business Management.
內容簡介：
Pharmacovigilance (drug safety signal detection or post-marketing surveillance) is an important research issue, both in the academic and practices, because postapproval adverse drug events (ADEs) are a global public health problem. Drug safety signal detection has traditionally relied on spontaneous reporting systems (SRSs), which contain reports of suspected adverse drug events collected from healthcare practitioners, patients, and pharmaceutical companies. Because SRSs are based on voluntary reporting, SRS-based methods incur several limitations including underreporting, overreporting, and misattribution of causality in drug-event combinations. To address the limitations of SRS-based systems, several research initiatives have explored the use of electronic health records (EHRs) (e.g., medical records databases or administrative/claims databases) for developing active surveillance systems. In this talk, I will present our study that develops an advanced analytics technique to detect adverse drug events (ADEs) from EHRs. Specifically, our proposed drug safety signal detection technique follows the learning to rank approach that learns from a collection of ranked drug-outcome pairs a signal ranking model, which can be employed to rank candidate signals relevant to a detection target. We employ the National Health Insurance Research Database (NHIRD) in Taiwan, a national-based insurance claim database, as our data source to develop and evaluate our proposed drug safety signal detection technique. I will discuss some important evaluation results.
（主辦：管理工程系）
