UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

A multiple hold-out framework for Sparse Partial Least Squares

Monteiro, JM; Rao, A; Shawe-Taylor, J; Mourao-Miranda, J; (2016) A multiple hold-out framework for Sparse Partial Least Squares. Journal of Neuroscience Methods , 271 pp. 182-194. 10.1016/j.jneumeth.2016.06.011. Green open access

[thumbnail of 1-s2.0-S0165027016301327-main.pdf]
Preview
Text
1-s2.0-S0165027016301327-main.pdf

Download (2MB) | Preview

Abstract

Background Supervised classification machine learning algorithms may have limitations when studying brain diseases with heterogeneous populations, as the labels might be unreliable. More exploratory approaches, such as Sparse Partial Least Squares (SPLS), may provide insights into the brain's mechanisms by finding relationships between neuroimaging and clinical/demographic data. The identification of these relationships has the potential to improve the current understanding of disease mechanisms, refine clinical assessment tools, and stratify patients. SPLS finds multivariate associative effects in the data by computing pairs of sparse weight vectors, where each pair is used to remove its corresponding associative effect from the data by matrix deflation, before computing additional pairs. New method We propose a novel SPLS framework which selects the adequate number of voxels and clinical variables to describe each associative effect, and tests their reliability by fitting the model to different splits of the data. As a proof of concept, the approach was applied to find associations between grey matter probability maps and individual items of the Mini-Mental State Examination (MMSE) in a clinical sample with various degrees of dementia. Results The framework found two statistically significant associative effects between subsets of brain voxels and subsets of the questions/tasks. Comparison with existing methods SPLS was compared with its non-sparse version (PLS). The use of projection deflation versus a classical PLS deflation was also tested in both PLS and SPLS. Conclusions SPLS outperformed PLS, finding statistically significant effects and providing higher correlation values in hold-out data. Moreover, projection deflation provided better results.

Type: Article
Title: A multiple hold-out framework for Sparse Partial Least Squares
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.jneumeth.2016.06.011
Publisher version: http://dx.doi.org/10.1016/j.jneumeth.2016.06.011
Language: English
Additional information: © 2016 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Keywords: Science & Technology, Life Sciences & Biomedicine, Biochemical Research Methods, Neurosciences, Biochemistry & Molecular Biology, Neurosciences & Neurology, Machine learning, Sparse methods, Partial Least Squares, Neuroimaging, Mini-Mental State Examination, Dementia, CANONICAL CORRELATION-ANALYSIS, SUPPORT VECTOR MACHINE, ALZHEIMERS-DISEASE, VARIABLE SELECTION, SEMANTIC DEMENTIA, BRAIN, CLASSIFICATION, ATROPHY, PATTERNS, REGULARIZATION
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI: https://discovery.ucl.ac.uk/id/eprint/1502727
Downloads since deposit
107Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item