Seminar Series: Keith Levin

Image
Kevin Lowres
April 8, 2021
3:00PM - 4:00PM
Location
Virtual Presentation

Date Range
Add to Calendar 2021-04-08 15:00:00 2021-04-08 16:00:00 Seminar Series: Keith Levin Title Averaging Connectomes: Beyond the Arithmetic Mean Meeting Link Speaker Keith Levin - University of Wisconsin - Madison, Department of Statistics Abstract Data arising from neuroimaging studies often consists of a collection of sample covariance matrices describing the dependence among blood oxygen level signals measured at different locations in the brain. A natural approach to population-level inference based on these samples is to consider the arithmetic mean of these matrices. However, the nature of the data under study suggests that this is a suboptimal choice, as the arithmetic mean fails to account for the structure of the positive definite cone. Even in the absence of covariance structure, the observed covariance matrices may differ in their noise structures owing to subject-level factors (e.g., head movement in the fMRI machine), in which case a weighted average is more appropriate. In this talk, we will discuss both these settings and present alternative choices of matrix averages better suited to them than the arithmetic mean. We will demonstrate some of these techniques in an application to fMRI data. Virtual Presentation Department of Statistics webmaster@stat.osu.edu America/New_York public
Description

Title

Averaging Connectomes: Beyond the Arithmetic Mean

Meeting Link

Speaker

Keith Levin - University of Wisconsin - Madison, Department of Statistics

Abstract

Data arising from neuroimaging studies often consists of a collection of sample covariance matrices describing the dependence among blood oxygen level signals measured at different locations in the brain. A natural approach to population-level inference based on these samples is to consider the arithmetic mean of these matrices. However, the nature of the data under study suggests that this is a suboptimal choice, as the arithmetic mean fails to account for the structure of the positive definite cone. Even in the absence of covariance structure, the observed covariance matrices may differ in their noise structures owing to subject-level factors (e.g., head movement in the fMRI machine), in which case a weighted average is more appropriate. In this talk, we will discuss both these settings and present alternative choices of matrix averages better suited to them than the arithmetic mean. We will demonstrate some of these techniques in an application to fMRI data.