Cargando…

The pitfalls of using Gaussian Process Regression for normative modeling

Normative modeling, a group of methods used to quantify an individual’s deviation from some expected trajectory relative to observed variability around that trajectory, has been used to characterize subject heterogeneity. Gaussian Processes Regression includes an estimate of variable uncertainty acr...

Descripción completa

Detalles Bibliográficos
Autores principales: Xu, Bohan, Kuplicki, Rayus, Sen, Sandip, Paulus, Martin P.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8443061/
https://www.ncbi.nlm.nih.gov/pubmed/34525108
http://dx.doi.org/10.1371/journal.pone.0252108
_version_ 1783753115789426688
author Xu, Bohan
Kuplicki, Rayus
Sen, Sandip
Paulus, Martin P.
author_facet Xu, Bohan
Kuplicki, Rayus
Sen, Sandip
Paulus, Martin P.
author_sort Xu, Bohan
collection PubMed
description Normative modeling, a group of methods used to quantify an individual’s deviation from some expected trajectory relative to observed variability around that trajectory, has been used to characterize subject heterogeneity. Gaussian Processes Regression includes an estimate of variable uncertainty across the input domain, which at face value makes it an attractive method to normalize the cohort heterogeneity where the deviation between predicted value and true observation is divided by the derived uncertainty directly from Gaussian Processes Regression. However, we show that the uncertainty directly from Gaussian Processes Regression is irrelevant to the cohort heterogeneity in general.
format Online
Article
Text
id pubmed-8443061
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-84430612021-09-16 The pitfalls of using Gaussian Process Regression for normative modeling Xu, Bohan Kuplicki, Rayus Sen, Sandip Paulus, Martin P. PLoS One Research Article Normative modeling, a group of methods used to quantify an individual’s deviation from some expected trajectory relative to observed variability around that trajectory, has been used to characterize subject heterogeneity. Gaussian Processes Regression includes an estimate of variable uncertainty across the input domain, which at face value makes it an attractive method to normalize the cohort heterogeneity where the deviation between predicted value and true observation is divided by the derived uncertainty directly from Gaussian Processes Regression. However, we show that the uncertainty directly from Gaussian Processes Regression is irrelevant to the cohort heterogeneity in general. Public Library of Science 2021-09-15 /pmc/articles/PMC8443061/ /pubmed/34525108 http://dx.doi.org/10.1371/journal.pone.0252108 Text en © 2021 Xu et al https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Xu, Bohan
Kuplicki, Rayus
Sen, Sandip
Paulus, Martin P.
The pitfalls of using Gaussian Process Regression for normative modeling
title The pitfalls of using Gaussian Process Regression for normative modeling
title_full The pitfalls of using Gaussian Process Regression for normative modeling
title_fullStr The pitfalls of using Gaussian Process Regression for normative modeling
title_full_unstemmed The pitfalls of using Gaussian Process Regression for normative modeling
title_short The pitfalls of using Gaussian Process Regression for normative modeling
title_sort pitfalls of using gaussian process regression for normative modeling
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8443061/
https://www.ncbi.nlm.nih.gov/pubmed/34525108
http://dx.doi.org/10.1371/journal.pone.0252108
work_keys_str_mv AT xubohan thepitfallsofusinggaussianprocessregressionfornormativemodeling
AT kuplickirayus thepitfallsofusinggaussianprocessregressionfornormativemodeling
AT sensandip thepitfallsofusinggaussianprocessregressionfornormativemodeling
AT paulusmartinp thepitfallsofusinggaussianprocessregressionfornormativemodeling
AT xubohan pitfallsofusinggaussianprocessregressionfornormativemodeling
AT kuplickirayus pitfallsofusinggaussianprocessregressionfornormativemodeling
AT sensandip pitfallsofusinggaussianprocessregressionfornormativemodeling
AT paulusmartinp pitfallsofusinggaussianprocessregressionfornormativemodeling