Biomedical Data Science

BMED-6265

Introduction of multivariate statistical methods to model and analyze recorded data from physiological systems in biomedical engineering. Statistical approaches related to applied multivariate statistics, classification and regression. Associated linear methods include principal component analysis, Fisher discriminant analysis, partial least squares, canonical correlation analysis and their nonlinear counterparts. Descriptive tools include scatter diagrams, Hotelling’s T2 statistics and contribution plots. The course has a strong emphasis on biomedical applications and presents associated results from studies related to autism, tissue engineering and stress-strain modeling.

Past Term Data