Cargando…

Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers

Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to opti...

Descripción completa

Detalles Bibliográficos
Autores principales: Wiersch, Lisa, Friedrich, Patrick, Hamdan, Sami, Komeyer, Vera, Hoffstaedter, Felix, Patil, Kaustubh R., Eickhoff, Simon B., Weis, Susanne
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Cold Spring Harbor Laboratory 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10491190/
https://www.ncbi.nlm.nih.gov/pubmed/37693374
http://dx.doi.org/10.1101/2023.08.30.555495
_version_ 1785104010596843520
author Wiersch, Lisa
Friedrich, Patrick
Hamdan, Sami
Komeyer, Vera
Hoffstaedter, Felix
Patil, Kaustubh R.
Eickhoff, Simon B.
Weis, Susanne
author_facet Wiersch, Lisa
Friedrich, Patrick
Hamdan, Sami
Komeyer, Vera
Hoffstaedter, Felix
Patil, Kaustubh R.
Eickhoff, Simon B.
Weis, Susanne
author_sort Wiersch, Lisa
collection PubMed
description Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to optimize generalization performance. In the present study, we systematically investigated the generalization performance of sex classification models trained on the parcelwise connectivity profile of either single samples or a compound sample containing data from four different datasets. Generalization performance was quantified in terms of mean across-sample classification accuracy and spatial consistency of accurately classifying parcels. Our results indicate that generalization performance of pwCs trained on single dataset samples is dependent on the specific test samples. Certain datasets seem to “match” in the sense that classifiers trained on a sample from one dataset achieved a high accuracy when tested on the respected other one and vice versa. The pwC trained on the compound sample demonstrated overall highest generalization performance for all test samples, including one derived from a dataset not included in building the training samples. Thus, our results indicate that a big and heterogenous training sample comprising data of multiple datasets is best suited to achieve generalizable results.
format Online
Article
Text
id pubmed-10491190
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Cold Spring Harbor Laboratory
record_format MEDLINE/PubMed
spelling pubmed-104911902023-09-09 Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers Wiersch, Lisa Friedrich, Patrick Hamdan, Sami Komeyer, Vera Hoffstaedter, Felix Patil, Kaustubh R. Eickhoff, Simon B. Weis, Susanne bioRxiv Article Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to optimize generalization performance. In the present study, we systematically investigated the generalization performance of sex classification models trained on the parcelwise connectivity profile of either single samples or a compound sample containing data from four different datasets. Generalization performance was quantified in terms of mean across-sample classification accuracy and spatial consistency of accurately classifying parcels. Our results indicate that generalization performance of pwCs trained on single dataset samples is dependent on the specific test samples. Certain datasets seem to “match” in the sense that classifiers trained on a sample from one dataset achieved a high accuracy when tested on the respected other one and vice versa. The pwC trained on the compound sample demonstrated overall highest generalization performance for all test samples, including one derived from a dataset not included in building the training samples. Thus, our results indicate that a big and heterogenous training sample comprising data of multiple datasets is best suited to achieve generalizable results. Cold Spring Harbor Laboratory 2023-09-01 /pmc/articles/PMC10491190/ /pubmed/37693374 http://dx.doi.org/10.1101/2023.08.30.555495 Text en https://creativecommons.org/licenses/by-nc-nd/4.0/This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/) , which allows reusers to copy and distribute the material in any medium or format in unadapted form only, for noncommercial purposes only, and only so long as attribution is given to the creator.
spellingShingle Article
Wiersch, Lisa
Friedrich, Patrick
Hamdan, Sami
Komeyer, Vera
Hoffstaedter, Felix
Patil, Kaustubh R.
Eickhoff, Simon B.
Weis, Susanne
Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title_full Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title_fullStr Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title_full_unstemmed Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title_short Sex classification from functional brain connectivity: Generalization to multiple datasets: Generalizability of sex classifiers
title_sort sex classification from functional brain connectivity: generalization to multiple datasets: generalizability of sex classifiers
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10491190/
https://www.ncbi.nlm.nih.gov/pubmed/37693374
http://dx.doi.org/10.1101/2023.08.30.555495
work_keys_str_mv AT wierschlisa sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT friedrichpatrick sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT hamdansami sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT komeyervera sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT hoffstaedterfelix sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT patilkaustubhr sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT eickhoffsimonb sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers
AT weissusanne sexclassificationfromfunctionalbrainconnectivitygeneralizationtomultipledatasetsgeneralizabilityofsexclassifiers