[Colloq] ECE Seminar Speaker, Jimeng Sun, on Oct. 31, 2014 (Friday), 10:30 am, 442 Dana

[Biron, Jessica]

Speaker: Jimeng Sun
Date: Oct. 31, 2014 (Friday)
Time: 10:30-11:30 am
Room: 442 Dana


Title:  Do it Once, Do it Right - Building a Scalable Predictive Modeling Platform for Healthcare Applications

Abstract: Predictive models are designed to predict the likelihood of one or more outcomes and are playing an increasing important role in biomedical research. Thanks to the explosion of Electronic Heart Records (EHR), the interest in building predictive models based on EHR data has skyrocketed in recent years.  There are some major challenges that remain to be addressed.  In this talk I will explore two of them.

• Effective algorithms are lacking in dealing with high-dimensional, longitudinal, sparse, inaccurate and inconsistent EHR data;
• The methodologies to develop predictive models are still labor intensive and ad-hoc. These rudimentary approaches are hindering the quality and throughput of healthcare and biomedical research.

In this talk, we promote a holistic approach that addresses both challenges by combining 1) algorithm development and 2) system building. We believe that a more robust and domain specific big-data platform could significantly speedup the development of robust and accurate predictive models for biomedical research.

I will present different projects covering both aspects of such a platform:

Algorithms: I will first describe our work on computational phenotyping from EHR data using sparse tensor factorization; then I will present a patient similarity method using supervised distance metric learning

System: I will introduce a parallel predictive modeling platform using Hadoop for enabling large scale modeling and exploration of big healthcare data

Bio: Jimeng Sun is an Associate Professor of School of Computational Science and Engineering at College of Computing in Georgia Institute of Technology. Prior to joining Georgia Tech, he was a research staff member at IBM TJ Watson Research Center. His research focuses on health analytics using electronic health records and data mining, especially in designing novel tensor analysis and similarity learning methods and developing large-scale predictive modeling systems.

Dr. Sun has worked on various healthcare applications such as computational phenotyping from electronic health records, heart failure onset prediction and hypertension control management. He has collaborated with many healthcare institutions including Vanderbilt university medical center, Children's healthcare of Atlanta, Center for Disease Control and Prevention (CDC), Geisinger Health System and Sutter Health.

He has published over 70 papers, filed over 20 patents (5 granted). He has received ICDM best research paper award in 2008, SDM best research paper award in 2007, and KDD Dissertation runner-up award in 2008. Dr. Sun received his B.S. and M.Phil. in Computer Science from Hong Kong University of Science and Technology in 2002 and 2003, and a PhD in Computer Science from Carnegie Mellon University in 2007.