Seminar at 2 pm today in 8th floor conference room of Bldg. 38A.

"Using Machine Learning and Statistical Methods to Predict Dire Outcomes Of  Patients With Community 
Acquired Pneumonia"

Constantin F. Aliferis M.D., Ph.D.
Assistant Professor, Division of Biomedical Informatics
Vanderbilt University

Abstract

Community-acquired pneumonia (CAP) is a major cause of death and lost  productivity in this country and 
also has an estimated annual cost of $4  billion. An important decision, both in terms of quality of care and 
resource  utilization, is whether individual patients should be hospitalised or treated at  home.

The optimality of this admission decision is predicated upon our ability to  successfully model the severity 
of each case and, more specifically, to  accurately predict its outcome.

In this talk, I will present results from two experiments exploring multiple  state-of-the-art machine 
learning and statistical methods to model mortality and  other adverse outcomes in patients with CAP. 
Some interesting methodological and  clinical/research implications and possibilities that follow from the 
above  results will also be discussed.

At the end of the talk I will provide a brief overview of the activities of the  Division of Biomedical 
Informatics of Vanderbilt University.