Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers

Abstract Depressed mood and anhedonia, the core symptoms of major depressive disorder (MDD), are linked to dysfunction in the brain’s reward and emotion regulation circuits. To develop a predictive model for treatment remission in MDD based on pre-treatment neurocircuitry and clinical features. A to...

Full description

Saved in:
Bibliographic Details
Main Authors: Rui Liu, Ximan Hou, Shuyu Liu, Yuan Zhou, Jingjing Zhou, Kaini Qiao, Han Qi, Ruinan Li, Zhi Yang, Ling Zhang, Jian Cui, Cheng Jin, Aihong Yu, Gang Wang
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01912-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849342067566706688
author Rui Liu
Ximan Hou
Shuyu Liu
Yuan Zhou
Jingjing Zhou
Kaini Qiao
Han Qi
Ruinan Li
Zhi Yang
Ling Zhang
Jian Cui
Cheng Jin
Aihong Yu
Gang Wang
author_facet Rui Liu
Ximan Hou
Shuyu Liu
Yuan Zhou
Jingjing Zhou
Kaini Qiao
Han Qi
Ruinan Li
Zhi Yang
Ling Zhang
Jian Cui
Cheng Jin
Aihong Yu
Gang Wang
author_sort Rui Liu
collection DOAJ
description Abstract Depressed mood and anhedonia, the core symptoms of major depressive disorder (MDD), are linked to dysfunction in the brain’s reward and emotion regulation circuits. To develop a predictive model for treatment remission in MDD based on pre-treatment neurocircuitry and clinical features. A total of 279 untreated MDD patients were analyzed, treated with selective serotonin reuptake inhibitors for 8–12 weeks, and assigned to training, internal validation, and external validation datasets. A hierarchical local-global imaging and clinical feature fusion graph neural network model was constructed. The model achieved 76.21% accuracy (AUC = 0.78) in predicting remission. Validation on the internal and external independent datasets yielded similar performance (accuracy = 72.73%, AUC = 0.74; accuracy = 71.43%, AUC = 0.72). Key contributing brain regions included the right globus pallidus, bilateral putamen, left hippocampus, bilateral thalamus, and bilateral anterior cingulate gyrus. These findings highlight the role of specific circuits in guiding antidepressant treatment.
format Article
id doaj-art-63aa6374cb144bc4bd64b73669a8e077
institution Kabale University
issn 2398-6352
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-63aa6374cb144bc4bd64b73669a8e0772025-08-20T03:43:30ZengNature Portfolionpj Digital Medicine2398-63522025-08-018111410.1038/s41746-025-01912-8Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkersRui Liu0Ximan Hou1Shuyu Liu2Yuan Zhou3Jingjing Zhou4Kaini Qiao5Han Qi6Ruinan Li7Zhi Yang8Ling Zhang9Jian Cui10Cheng Jin11Aihong Yu12Gang Wang13Beijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityMedical Robot Research Institute, School of Biomedical Engineering, Shanghai Jiao Tong UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityDepartment of Psychiatry, Shandong Daizhuang HospitalBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityBeijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical UniversityAbstract Depressed mood and anhedonia, the core symptoms of major depressive disorder (MDD), are linked to dysfunction in the brain’s reward and emotion regulation circuits. To develop a predictive model for treatment remission in MDD based on pre-treatment neurocircuitry and clinical features. A total of 279 untreated MDD patients were analyzed, treated with selective serotonin reuptake inhibitors for 8–12 weeks, and assigned to training, internal validation, and external validation datasets. A hierarchical local-global imaging and clinical feature fusion graph neural network model was constructed. The model achieved 76.21% accuracy (AUC = 0.78) in predicting remission. Validation on the internal and external independent datasets yielded similar performance (accuracy = 72.73%, AUC = 0.74; accuracy = 71.43%, AUC = 0.72). Key contributing brain regions included the right globus pallidus, bilateral putamen, left hippocampus, bilateral thalamus, and bilateral anterior cingulate gyrus. These findings highlight the role of specific circuits in guiding antidepressant treatment.https://doi.org/10.1038/s41746-025-01912-8
spellingShingle Rui Liu
Ximan Hou
Shuyu Liu
Yuan Zhou
Jingjing Zhou
Kaini Qiao
Han Qi
Ruinan Li
Zhi Yang
Ling Zhang
Jian Cui
Cheng Jin
Aihong Yu
Gang Wang
Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
npj Digital Medicine
title Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
title_full Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
title_fullStr Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
title_full_unstemmed Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
title_short Predicting antidepressant response via local-global graph neural network and neuroimaging biomarkers
title_sort predicting antidepressant response via local global graph neural network and neuroimaging biomarkers
url https://doi.org/10.1038/s41746-025-01912-8
work_keys_str_mv AT ruiliu predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT ximanhou predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT shuyuliu predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT yuanzhou predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT jingjingzhou predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT kainiqiao predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT hanqi predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT ruinanli predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT zhiyang predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT lingzhang predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT jiancui predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT chengjin predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT aihongyu predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers
AT gangwang predictingantidepressantresponsevialocalglobalgraphneuralnetworkandneuroimagingbiomarkers