Discovery Research with Electronic Medical Records Data

• Activity Number: 581
• Type: Invited
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Tianxi Cai*
• Companies: Harvard University
• Keywords: phenotyping ; classification ; EMR ; data integration ; semi-supervised learning

Abstract

In clinical practice, patients with the same disease diagnosis often differ in outcomes and response to treatment. The ability to both classify and predict disease phenotypes would be a valuable asset in clinical decision-making. Large datasets containing both a wealth of clinical and experimental data now exist as a result of the increasing adoption of electronic medical records (EMR) linked with specimen bio-repositories. These datasets allow for data driven classification and prediction of sub-phenotypes and investigation of shared risk factors across a group of phenotypes. In this talk, I'll discuss various statistical methods that illustrate both the challenges and potential opportunities that arise from analyzing EMR data.