{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T12:00:20Z","timestamp":1790164820787,"version":"4.1.0"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Psychiatry"],"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To determine whether combining resting-state and the verbal fluency task (VFT) prefrontal fNIRS features with machine learning can distinguish patients with MDD from healthy controls (HCs), and to clarify the relative contributions of functional connectivity (FC) versus amplitude features to classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We enrolled 98 patients with MDD and 51 HCs, matched on age, sex, education, and handedness. Participants completed a resting-state fNIRS recording and a VFT using a multichannel prefrontal\u2013temporal montage. After preprocessing, oxyhemoglobin (HbO)\u2013based amplitude features and Pearson correlation\u2013based FC features were extracted from two conditions. Five classifiers\u2014logistic regression, k-nearest neighbors, support vector machine (SVM), partial least squares, and linear discriminant analysis\u2014were evaluated using nested stratified cross-validation. Model performance was quantified using out-of-fold (OOF) predicted probabilities from the outer loop and summarized by accuracy, sensitivity, specificity, and area under the ROC curve (AUC). The 95% confidence intervals (95% CIs) for AUC were estimated using DeLong\u2019s method. The classification threshold was determined using the Youden index on the OOF ROC curve. SHAP analysis was then applied to the best-performing model to quantify the contribution of each feature to the predicted probability of MDD.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Demographic variables did not differ significantly between groups. Across all classifiers, the SVM achieved the best discrimination between MDD and HCs, with OOF AUC of 0.867 (95% CI: 0.805\u20130.928) and an accuracy of 0.839 (95% CI: 0.770\u20130.894). At the Youden-optimal threshold (0.482), the SVM model yielded a sensitivity of 0.918 (95% CI: 0.845\u20130.964) and a specificity of 0.686 (95% CI: 0.541\u20130.809). SHAP analysis indicated that the top-ranked predictors were predominantly FC features from both resting-state and VFT conditions, particularly IPFC\u2013TPC connectivity at rest and FPC\u2013TPC and IPFC\u2013DLPFC connectivity during VFT, whereas regional HbO amplitude measures contributed less.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The present study suggests that network-level dysconnectivity, which fNIRS captures, plays a central role in distinguishing MDD from HCs, rather than isolated regional hypoactivation. With its high sensitivity, low cost, and SHAP-based interpretability, the SVM model provides an interpretable framework for investigating fNIRS-based neurophysiological patterns associated with MDD and may serve as a potential adjunctive approach for future screening applications after external validation. Multi-center external validation is required before routine implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpsyt.2026.1767469","type":"journal-article","created":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T11:06:41Z","timestamp":1790161601000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine learning classification of fNIRS signals in patients with major depressive disorder and healthy controls"],"prefix":"10.3389","volume":"17","author":[{"given":"Zhuoyan","family":"Shan","sequence":"first","affiliation":[{"name":"School of Mental Health, Jining Medical University","place":["Jining, Shandong, China"]},{"name":"Shandong Daizhuang Hospital","place":["Jining, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xia","family":"Liu","sequence":"additional","affiliation":[{"name":"Shandong Daizhuang Hospital","place":["Jining, Shandong, China"]},{"name":"Sleep Medicine Center of Shandong Daizhuang Hospital","place":["Jining, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyu","family":"Lin","sequence":"additional","affiliation":[{"name":"Shandong Daizhuang Hospital","place":["Jining, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hao","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Mental Health, Jining Medical University","place":["Jining, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shanling","family":"Ji","sequence":"additional","affiliation":[{"name":"School of Mental Health, Jining Medical University","place":["Jining, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,23]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1038\/s41392-024-01738-y","article-title":"Major depressive disorder: hypothesis, mechanism, prevention and treatment","volume":"9","author":"Cui","year":"2024","journal-title":"Signal Transd Targ Ther"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2022.104027","article-title":"Identifying neuroimaging biomarkers of major depressive disorder from cortical hemodynamic responses using machine learning approaches","volume":"79","author":"Li","year":"2022","journal-title":"EBioMedicine"},{"key":"B3","doi-asserted-by":"publisher","first-page":"21257","DOI":"10.1038\/s41598-025-05896-z","article-title":"Different prefrontal cortex activity patterns in bipolar and unipolar depression during verbal fluency tasks based on functional near infrared spectroscopy study","volume":"15","author":"Mou","year":"2025","journal-title":"Sci Rep"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2021.664859","article-title":"Abnormality of resting-state functional connectivity in major depressive disorder: A study with whole-head near-infrared spectroscopy","volume":"12","author":"Sakakibara","year":"2021","journal-title":"Front Psychiatry"},{"key":"B5","doi-asserted-by":"publisher","first-page":"83","DOI":"10.1016\/j.jad.2018.01.010","article-title":"Hypofunction of left dorsolateral prefrontal cortex in depression during verbal fluency task: A multi-channel near-infrared spectroscopy study","volume":"231","author":"Akiyama","year":"2018","journal-title":"J Affect Disord"},{"key":"B6","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1016\/j.jad.2019.04.001","article-title":"Brain functional alterations observed 4-weekly in major depressive disorder following antidepressant treatment","volume":"252","author":"Yamagata","year":"2019","journal-title":"J Affect Disord"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2023.1235713","article-title":"Associations between frontal lobe activity and depressive symptoms in patients with major depressive disorder receiving rTMS treatment: a near-infrared spectroscopy study","volume":"14","author":"Chou","year":"2023","journal-title":"Front Psychiatry"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1109\/jbhi.2021.3076762","article-title":"Depression analysis and recognition based on functional near-infrared spectroscopy","volume":"25","author":"Wang","year":"2021","journal-title":"IEEE J BioMed Health Inform"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.3389\/fnrgo.2023.994969","article-title":"Benchmarking framework for machine learning classification from fNIRS data","volume":"4","author":"Benerradi","year":"2023","journal-title":"Front Neuroergon"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1117\/1.nph.11.2.025001","article-title":"Functional near-infrared spectroscopy-based diagnosis support system for distinguishing between mild and severe depression using machine learning approaches","volume":"11","author":"Huang","year":"2024","journal-title":"Neurophotonics"},{"key":"B11","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.jad.2024.08.082","article-title":"Identifying neuroimaging biomarkers in major depressive disorder using machine learning algorithms and functional near-infrared spectroscopy (fNIRS) during verbal fluency task","volume":"365","author":"Mao","year":"2024","journal-title":"J Affect Disord"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajp.2023.103901","article-title":"Interpretable deep learning model for major depressive disorder assessment based on functional near-infrared spectroscopy","volume":"92","author":"Ho","year":"2024","journal-title":"Asian J Psychiatr"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2020.00724","article-title":"Functional near-infrared spectroscopy and its clinical application in the field of neuroscience: Advances and future directions","volume":"14","author":"Chen","year":"2020","journal-title":"Front Neurosci"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpsychires.2021.06.015","article-title":"Probing depression, schizophrenia, and other psychiatric disorders using fNIRS and the verbal fluency test: A systematic review and meta-analysis","volume":"140","author":"Yeung","year":"2021","journal-title":"J Psychiatr Res"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajp.2025.104382","article-title":"Functional near-infrared spectroscopy (fNIRS) in patients with major depressive disorder, generalized anxiety disorder and their comorbidity: Comparison with healthy controls","volume":"105","author":"Shen","year":"2025","journal-title":"Asian J Psychiatr"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroscience.2025.07.012","article-title":"The association between resting-state prefrontal network and cognitive function in depressed patients with insomnia","volume":"581","author":"Lin","year":"2025","journal-title":"Neuroscience"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1177\/1073858413502707","article-title":"Resting-state functional brain connectivity: lessons from functional near-infrared spectroscopy","volume":"20","author":"Niu","year":"2014","journal-title":"Neuroscientist"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.bbr.2024.115218","article-title":"Neural mechanisms underlying cognitive impairment in depression and cognitive benefits of exercise intervention","volume":"476","author":"Tian","year":"2025","journal-title":"Behav Brain Res"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2024.08.022","article-title":"Analysis of functional network asymmetry in major depressive disorder under four fNIRS tasks","volume":"365","author":"Yang","year":"2024","journal-title":"J Affect Disord"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajp.2025.104445","article-title":"Structural alterations after repetitive transcranial magnetic stimulation in depression and the link to neurotransmitter profiles","volume":"107","author":"Ping","year":"2025","journal-title":"Asian J Psychiatr"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1109\/tnsre.2021.3115266","article-title":"fNIRS evidence for distinguishing patients with major depression and healthy controls","volume":"29","author":"Chao","year":"2021","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2024.01.044","article-title":"Differences in prefrontal cortex activation in Chinese college students with different severities of depressive symptoms: A large sample of functional near-infrared spectroscopy (fNIRS) findings","volume":"350","author":"Wu","year":"2024","journal-title":"J Affect Disord"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1017\/s0033291725000303","article-title":"Resting-state network alterations in depression: a comprehensive meta-analysis of functional connectivity","volume":"55","author":"Zhang","year":"2025","journal-title":"Psychol Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1016\/j.pnpbp.2017.05.001","article-title":"Reduced language lateralization in first episode schizophrenia: A near infrared spectroscopy study","volume":"78","author":"Chou","year":"2017","journal-title":"Prog Neuro-Psychopharmacol Biol Psychiatry"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2023.1205931","article-title":"Automatic depression diagnosis through hybrid EEG and near-infrared spectroscopy features using support vector machine","volume":"17","author":"Yi","year":"2023","journal-title":"Front Neurosci"}],"container-title":["Frontiers in Psychiatry"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1767469\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T11:06:42Z","timestamp":1790161602000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyt.2026.1767469\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,23]]},"references-count":25,"alternative-id":["10.3389\/fpsyt.2026.1767469"],"URL":"https:\/\/doi.org\/10.3389\/fpsyt.2026.1767469","relation":{},"ISSN":["1664-0640"],"issn-type":[{"value":"1664-0640","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,23]]},"article-number":"1767469"}}