Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks

IntroductionFunctional brain networks measured by resting-state functional magnetic resonance imaging (rs-fMRI) have become a promising tool for understanding the neural mechanisms underlying schizophrenia (SZ). However, the high dimensionality of these networks and small sample sizes pose significa...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinyan Yuan, Shaolong Wei, Ying Sun, Lingling Gu, Yanyan He, Tiantian Chen, Hongcheng Yao, Haonan Rao
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Neuroscience
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fnins.2025.1609547/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850096639646105600
author Xinyan Yuan
Shaolong Wei
Ying Sun
Lingling Gu
Yanyan He
Tiantian Chen
Hongcheng Yao
Haonan Rao
author_facet Xinyan Yuan
Shaolong Wei
Ying Sun
Lingling Gu
Yanyan He
Tiantian Chen
Hongcheng Yao
Haonan Rao
author_sort Xinyan Yuan
collection DOAJ
description IntroductionFunctional brain networks measured by resting-state functional magnetic resonance imaging (rs-fMRI) have become a promising tool for understanding the neural mechanisms underlying schizophrenia (SZ). However, the high dimensionality of these networks and small sample sizes pose significant challenges for effective classification and model generalization.MethodsWe propose a robust multi-task feature selection method combined with counterfactual explanations to improve the accuracy and interpretability of SZ identification. rs-fMRI data are preprocessed to construct a functional connectivity matrix, and features are extracted by sorting the upper triangular elements. A multi-task feature selection framework based on the Gray Wolf Optimizer (GWO) is developed to identify abnormal functional connectivity (FC) features in SZ patients. A counterfactual explanation model is applied to reduce perturbations in abnormal FC features, returning the model prediction to normal and enhancing clinical interpretability.ResultsOur method was tested on five real-world SZ datasets. The results demonstrate that the proposed method significantly outperforms existing methods in terms of classification accuracy while offering new insights into the analysis of SZ through improved feature selection and explanation.DiscussionThe integration of multi-task feature selection and counterfactual explanation improves both the accuracy and interpretability of SZ identification. This approach provides valuable clinical insights by revealing the key functional connectivity features associated with SZ, which could assist in the development of more effective diagnostic tools.
format Article
id doaj-art-4ccd0d7dd35447519c7cbe2b6710f430
institution DOAJ
issn 1662-453X
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neuroscience
spelling doaj-art-4ccd0d7dd35447519c7cbe2b6710f4302025-08-20T02:41:11ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2025-07-011910.3389/fnins.2025.16095471609547Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networksXinyan Yuan0Shaolong Wei1Ying Sun2Lingling Gu3Yanyan He4Tiantian Chen5Hongcheng Yao6Haonan Rao7School of Electronics and Information, Jiangsu Vocational College of Business, Nantong, ChinaSchool of Artificial Intelligence and Computer Science, Nantong University, Nantong, ChinaSchool of Electronics and Information, Jiangsu Vocational College of Business, Nantong, ChinaSchool of Electronics and Information, Jiangsu Vocational College of Business, Nantong, ChinaSchool of Electronics and Information, Jiangsu Vocational College of Business, Nantong, ChinaSchool of Electronics and Information, Jiangsu Vocational College of Business, Nantong, ChinaSchool of Information Science and Technology, Nantong University, Nantong, ChinaSchool of Information Science and Technology, Nantong University, Nantong, ChinaIntroductionFunctional brain networks measured by resting-state functional magnetic resonance imaging (rs-fMRI) have become a promising tool for understanding the neural mechanisms underlying schizophrenia (SZ). However, the high dimensionality of these networks and small sample sizes pose significant challenges for effective classification and model generalization.MethodsWe propose a robust multi-task feature selection method combined with counterfactual explanations to improve the accuracy and interpretability of SZ identification. rs-fMRI data are preprocessed to construct a functional connectivity matrix, and features are extracted by sorting the upper triangular elements. A multi-task feature selection framework based on the Gray Wolf Optimizer (GWO) is developed to identify abnormal functional connectivity (FC) features in SZ patients. A counterfactual explanation model is applied to reduce perturbations in abnormal FC features, returning the model prediction to normal and enhancing clinical interpretability.ResultsOur method was tested on five real-world SZ datasets. The results demonstrate that the proposed method significantly outperforms existing methods in terms of classification accuracy while offering new insights into the analysis of SZ through improved feature selection and explanation.DiscussionThe integration of multi-task feature selection and counterfactual explanation improves both the accuracy and interpretability of SZ identification. This approach provides valuable clinical insights by revealing the key functional connectivity features associated with SZ, which could assist in the development of more effective diagnostic tools.https://www.frontiersin.org/articles/10.3389/fnins.2025.1609547/fullschizophreniafunctional connectivityrs-fMRIfeature selectioncounterfactual explanation
spellingShingle Xinyan Yuan
Shaolong Wei
Ying Sun
Lingling Gu
Yanyan He
Tiantian Chen
Hongcheng Yao
Haonan Rao
Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
Frontiers in Neuroscience
schizophrenia
functional connectivity
rs-fMRI
feature selection
counterfactual explanation
title Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
title_full Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
title_fullStr Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
title_full_unstemmed Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
title_short Robust multi-task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
title_sort robust multi task feature selection with counterfactual explanation for schizophrenia identification using functional brain networks
topic schizophrenia
functional connectivity
rs-fMRI
feature selection
counterfactual explanation
url https://www.frontiersin.org/articles/10.3389/fnins.2025.1609547/full
work_keys_str_mv AT xinyanyuan robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT shaolongwei robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT yingsun robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT linglinggu robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT yanyanhe robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT tiantianchen robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT hongchengyao robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks
AT haonanrao robustmultitaskfeatureselectionwithcounterfactualexplanationforschizophreniaidentificationusingfunctionalbrainnetworks