35.458, FYI: Online lecture 2/15, Alex Teghipco: predicting language impairment from neuroimaging data

[The LINGUIST List]

LINGUIST List: Vol-35-458. Thu Feb 08 2024. ISSN: 1069 - 4875.

Subject: 35.458, FYI: Online lecture 2/15, Alex Teghipco: predicting language impairment from neuroimaging data

Moderators: Malgorzata E. Cavar, Francis Tyers (linguist at linguistlist.org)
Managing Editor: Justin Fuller
Team: Helen Aristar-Dry, Steven Franks, Everett Green, Daniel Swanson, Maria Lucero Guillen Puon, Zackary Leech, Lynzie Coburn, Natasha Singh, Erin Steitz
Jobs: jobs at linguistlist.org | Conferences: callconf at linguistlist.org | Pubs: pubs at linguistlist.org

Homepage: http://linguistlist.org

Please support the LL editors and operation with a donation at:
           https://funddrive.linguistlist.org/donate/

Editor for this issue: Justin Fuller <justin at linguistlist.org>
================================================================


Date: 07-Feb-2024
From: Dirk Den Ouden [denouden at sc.edu]
Subject: Online lecture 2/15, Alex Teghipco: predicting language impairment from neuroimaging data


Thursday, February 15th, 12pm EST (5pm UTC)
Presentation in Zoom, accessible via the C-STAR website:
http://cstar.sc.edu/lecture-series/

Enhancing prediction of language impairment from neuroimaging data

Alex Teghipco, Ph.D.
University of South Carolina

Machine learning stands to revolutionize our understanding of
neurobiology by exploiting individual differences, representing a
radical departure from the traditional approach of smoothing out
individual variations to make more generalized conclusions about
populations. While this approach has enjoyed success in neuroimaging
and holds promise in clinical settings, where personalized predictions
stand to improve patient management, it tends to demand massive sample
sizes. Here, I present evidence that current structural neuroimaging
datasets in chronic stroke are sufficient in size to begin
capitalizing on a previously unmapped source of variance in
aphasia—spatial information about features embedded in a multivariate
space, which can be identified with 3D Convolutional Neural Networks
and used to improve prediction of aphasia severity. Despite the
potential for unraveling more complex patterns of brain structure and
function, machine learning with even relatively simple algorithms can
be difficult in the average clinical neuroimaging study, particularly
if limited to functional data, because sample sizes remain small,
risking models to overfit to noise. To address this challenge, I
introduce a pipeline for classical machine learning (e.g., with
Support Vector Machines) that improves prediction of individual
language outcomes by identifying reliably predictive features to
model, framing the challenge of finding important features as the
starting point of the analysis rather than the end. This pipeline is
first validated with lesion symptom mapping in a large sample, then
shown to outperform other pipelines in smaller functional neuroimaging
datasets (CT perfusion), enabling prediction of language impairment in
the average clinical neuroimaging study.

_______________________________________________

The online lecture can be followed online from your computer, tablet
or smartphone, in Zoom. The zoom link is accessible via the C-STAR
website: http://cstar.sc.edu/lecture-series/

The live in-person lecture will be in Discovery I room #140 (915
Greene Street, Columbia, SC)

For more information, or to be added to the C-STAR mailing list,
contact Dirk den Ouden: denouden at sc.edu

Linguistic Field(s): Applied Linguistics
                     Clinical Linguistics
                     Cognitive Science
                     Neurolinguistics




------------------------------------------------------------------------------

Please consider donating to the Linguist List https://give.myiu.org/iu-bloomington/I320011968.html


LINGUIST List is supported by the following publishers:

John Benjamins http://www.benjamins.com/

Lincom GmbH https://lincom-shop.eu/

Linguistic Association of Finland http://www.ling.helsinki.fi/sky/

Multilingual Matters http://www.multilingual-matters.com/

Netherlands Graduate School of Linguistics / Landelijke (LOT) http://www.lotpublications.nl/

Wiley http://www.wiley.com


----------------------------------------------------------
LINGUIST List: Vol-35-458
----------------------------------------------------------