2016～2017学年第一学期第七周学术活动安排1

欢迎广大师生前往！

                                                                                                                             校科协、计算机学院

                                                                                                            2016年10月8日

报告题目及摘要、专家简介

1. 医学知识图谱及其在临床决策支持中的应用

黄智生，荷兰阿姆斯特丹自由大学教授

Abstract：传统的医疗模式经常是通过医生的个人经验积累来提高诊疗水平。这种方式现在已经是不值得鼓励的，因为医学经验的积累通常是以病人的痛苦（甚至生命）和医疗资源的某种浪费为代价的。大数据时代已经为我们提供了极其丰富的医学知识资源。在此基础上所构建的医学知识图谱为医学知识的共享与传播，诊疗手段的规范化，及其临床决策科学化提供了良好的环境和技术基础。本报告将通过一系列具体的实例介绍医学知识图谱在临床决策支持中的应用，并阐述医学知识图谱广泛的发展前景和应用价值。

Bio：黄智生博士，荷兰阿姆斯特丹自由大学终身教授，中国江苏科技大学特聘教授，和js33333线路登录等五所大学兼职教授。出版了《海量语义数据处理-平台，技术，与应用》《生物医学语义技术》等教材，发表过论文论著超过二百篇，

担任超过一百个国际学术会议的程序委员会委员，超过二十个国际会议的联合主席，担任六个国际学术刊物的编委，特约主编或特约编委。主持欧盟第七框架重大项目LarKC中推理工作组的工作。主持欧盟第七框架智慧医疗重大项目EURECA中基于语义技术的临床试验系统 SemanticCT的开发；主持了基于语义技术的抗菌药物合理用药系统SeSRUA的开发。他参与开发的E-Culture项目在2006年世界语义万维网技术挑战赛上获得冠军。作为第一作者获得2014世界健康信息技术学术年会（HealthInfo2014）最佳论文奖。

2. DTMiner: Mining disease-gene association from biomedical literature

朱其立，上海交通大学 教授

Abstract: In this study, we propose a reliable and efficient framework that takes large biomedical literature repositories as inputs, identifies credible relationships between diseases and genes, and presents possible genes related to a given disease and possible diseases related to a given gene. The framework incorporates name entity recognition (NER), whichidentifies occurrences of genes and diseases in texts, association detection whereby we extract and evaluate features from gene-disease pairs, and ranking algorithms that estimate how closely the pairs are related. The F1-score of the NER phase is 0.87, which is higher than existing studies.

The association detection phase takes drastically less time than previous work while maintaining a comparable F1-score of 0.86. The end-to-end result achieves a 0.259 F1-score for the top 50 genes associated with a disease, which performs better than previous work. In addition, we released a web service for public use of the dataset.

Bio: Kenny Qili Zhu is a Professor of computer science at Department of Computer Science and Engineering at Shanghai Jiao Tong University. He graduated with B.Eng (Hons) in Electrical Engineering in 1999 and PhD in Computer Science in 2005 from National University of Singapore. He was a postdoctoral researcher and lecturer from 2007 to 2009 in the programming language group at Princeton University. Prior to that , he was a softwaredesign engineer in the Windows Live Identity group of Microsoft in Redmond, WA. From Feb 2010 to Aug 2010, he was a visiting professor at Microsoft Research Asia in Beijing. Kenny was the winner of the 2013 Google Faculty Research Award and DASFAA 2014 Best Paper Award.

3. 知识挖掘对中国分级诊疗的智能辅助

闫峻，微软亚洲研究院 博士，高级研究经理

Abstract: 中国的分级诊疗制度在推进过程中面临诸多困难。其中有两个大的挑战，它们分别是大量基层医生专业水平具有局限性以及基层医生数量不足导致的工作负担过重。这些都严重影响了分级诊疗的推广普及。为了通过技术手段辅助解决这两个问题，我们运用知识挖掘，自然语言理解以及信息检索技术协助基层医生更方便的获取临床知识指导，通过软件机器人有效降低基层医生重复劳动的工作负担。这个报告会重点介绍分级诊疗中的具体问题以及基于知识计算的解决方案。

Bio: Dr. Jun Yan received the Ph.D. degree in digital signal processing and pattern recognition from the department of information science, school of mathematical science, Peking University, P.R. China. During his Ph.D., he has been a research intern of MSRA from 2003 to 2005 and awarded as Microsoft fellow in 2004. Before join Microsoft, he has been a research associate at CBI, HMS, Harvard, Cambridge, MA, in 2005. He joined Microsoft Research Asia (MSRA) from 2006. Currently he is working in the Data Mining and Enterprise Intelligence group of MSRA as senior research manager. His research interests are on knowledge mining for AI, text data preprocessing,  information retrieval and behavior targeted online advertising etc. His current focus application is for healthcare. So far, he has successfully incubated tens of technologies, which have been used in Microsoft products. In academia, he has more than 60 quality papers published in referred conferences and journals, including SIGKDD, SIGIR, WWW, ICDM, TKDE, etc. He has been the PC members of international conferences SIGKDD, SIGIR etc. and is also reviewers of journals articles TKDE, TPAMI etc.

4. 中医药知识图谱的构建与应用

于彤，中国中医科学院中医药信息研究所助理研究员

Abstract: “中医药学语言系统”是以本体的技术理念构建的大型语言系统。它已发展为一个包含10余万个中医概念以及100余万个语义关系的大型语义网络，基本覆盖了中医药学科的概念体系。我们以中医药学语言系统为骨架，集成中医药领域现有的数据库，构成中医药知识图谱，可用于知识检索、知识可视化等应用。本报告首先简要介绍中医药应用背景，进而阐述中医药知识图谱的数据来源、构建方法和应用，最后讨论中医药知识图谱的发展前景。

Bio：于彤，2012年获浙江大学计算机博士学位，现为中国中医科学院中医药信息研究所助理研究员，在中医药信息学领域具有近十年的研究经历，主要研究兴趣包括：本体、语义网、知识工程等，参与研发了中医药学语言系统、中医临床术语系统、中医药知识服务平台以及中医药知识图谱。

5. 全科医生抗糖药物推荐系统研究

马敬东，华中科技大学同济医学院副教授

Bio:马敬东，博士、副教授、硕士生导师。现任华中科技大学同济医学院医药卫生管理学院信息系主任，院长助理。中国卫生信息学会理事，教育专业委员会常务委员，中华医学会医学信息学分会委员。曾任哈佛大学公共卫生学院Takemi研究员。研究兴趣包括电子健康管理，健康大数据分析。