Faculty of Science, Engineering & Technology

Predictive modelling of high-dimensional stroke data

Summary

Start Date

4th Dec 2012 2:00pm

End Date

4th Dec 2012 3:00pm

Predictive modelling of high-dimensional stroke data

Dr. Alan Dabney

Department of Statistics

Texas A&M University

2 pm December 4, Physics Lecture Theatre 2

When a stroke victim arrives at the emergency room, physicians must make rapid decisions to decide on the appropriate treatment. The wrong decision can easily lead to a lost life. Micro RNA (miRNA) are small molecules that regulate gene expression. Since the miRNA from various tissues end up in the blood and can be assayed easily, miRNA have the potential to be used as real-time diagnostic indicators of the health status of an individual. In collaboration with physicians at the St. Joseph hospital in Bryan, Texas, we are collecting a large cohort of human stroke victims, based on which we will create predictive models for stroke severity. In this talk I will discuss the statistical principles behind the predictive modelling of high-dimensional data, using the stroke data as a case study.

Seminar notice (pdf, 150KB)