[BIC-announce] Psych seminar: fMRI data-reduction techniques

[Sylvain Baillet, Dr]

From: Heungsun Hwang, Dr. [heungsun.hwang at mcgill.ca<mailto:heungsun.hwang at mcgill.ca>]
Sent: November 17, 2014 12:19 PM

Subject: Quant Brownbag

Dear all,

In this week’s quant brownbag, Ji Yeh Choi will present her recent work that combines two popular data-reduction techniques for functional data into a unified framework. She applied her approach for the analysis of fMRI data. The brownbag will be in STBIO W7/21 at 10:00 on Thursday. The title and abstract of her talk are given below.

Title: A unified approach to functional principal component analysis and functional multiple-set canonical correlation

Abstract:  Functional principal component analysis and functional multiple-set canonical correlation analysis are data reduction techniques for functional data that are assumed to arise from an underlying smooth function varying over a continuum such as time or space. In the former, low-dimensional components are extracted from a single set of functional data such that they explain the most variance of the dataset, whereas in the latter, low-dimensional components are obtained from each of multiple functional datasets in such a way that the associations among the components are maximized across the different sets. In this paper, we propose a unified approach to functional principal component analysis and functional multiple-set canonical correlation analysis. The proposed approach subsumes the two techniques as special cases. Also importantly, it can seek for a compromise between the techniques in estimating low-dimensional components, maximizing the associations among the components obtained from each set of functional data and at the same time accounting for the variance of the data well. Technically, the proposed approach can be viewed as a functional extension of Hwang et al.’s (2013) approach to combining principal component analysis and multiple-set canonical correlation analysis into a single framework. We propose a single optimization function for the proposed approach, and develop an alternating regularized least squares algorithm to estimate parameters in combination with basis-function approximations to functions. We apply the proposed approach for the analysis of multiple-subject functional magnetic resonance imaging data to obtain low-dimensional components of blood-oxygen level dependent signal changes over time, which are highly correlated across the subjects as well as representative of the data. The extracted components are used to identify networks of neural activity that are commonly activated across the subjects while carrying out a working memory task.
Ji Yeh Choi is a doctoral student in quantitative psychology at McGill.

Detailed information on the brownbag can be found here: http://www.psych.mcgill.ca/misc/quantsem/index.html

Thanks,
Heungsun

==============================================
Heungsun Hwang,  Associate Professor
Department of Psychology, McGill University
1205 Dr. Penfield Avenue, Montreal, QC H3A 1B1, Canada
Tel: 514-398-8021        Fax: 514-398-4896
www.psych.mcgill.ca/perpg/fac/hwang/hhwang.htm<http://www.psych.mcgill.ca/perpg/fac/hwang/hhwang.htm>
==============================================


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20141117/dabe2c14/attachment.html>