{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T16:40:48Z","timestamp":1792082448782,"version":"4.4.0"},"reference-count":62,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Ment Health"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Depression is one of the most prevalent mental health disorders globally, affecting approximately 280 million people and frequently going undiagnosed or misdiagnosed. The growing ubiquity of wearable devices enables continuous monitoring of activity levels, providing a new avenue for data-driven detection and severity assessment of depression. However, existing machine learning models often exhibit lower performance when distinguishing overlapping subtypes of depression and frequently lack explainability, an essential component for clinical acceptance.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and evaluate an interpretable machine learning framework for detecting depression and classifying its severity using wearable-actigraphy data, while addressing common challenges such as imbalanced datasets and limited model transparency.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n We used the Depresjon dataset and applied Adaptive Synthetic Sampling (ADASYN) to mitigate class imbalance. We extracted multiple statistical features (eg, power spectral density mean and autocorrelation) and demographic attributes (eg, age) from the raw activity data. Five machine learning algorithms (logistic regression, support vector machines, random forest, XGBoost, and neural networks) were assessed via accuracy, precision, recall,\n F<\/jats:italic>\n 1<\/jats:sub>\n -score, specificity, and Matthew correlation constant. We further used Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) to elucidate prediction drivers.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n XGBoost achieved the highest overall accuracy of 84.94% for binary classification and 85.91% for multiclass severity. SHAP and LIME revealed power spectral density mean, age, and autocorrelation as top predictors, highlighting circadian disruptions\u2019 role in depression.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Our interpretable framework reliably identifies depressed versus nondepressed individuals and differentiates mild from moderate depression. The inclusion of SHAP and LIME provides transparent, clinically meaningful insights, emphasizing the potential of explainable artificial intelligence to enhance early detection and intervention strategies in mental health care.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/72038","type":"journal-article","created":{"date-parts":[[2025,9,11]],"date-time":"2025-09-11T20:21:13Z","timestamp":1757622073000},"page":"e72038-e72038","source":"Crossref","is-referenced-by-count":37,"title":["Explainable AI for Depression Detection and Severity Classification From Activity Data: Development and Evaluation Study of an Interpretable Framework"],"prefix":"10.2196","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7863-3746","authenticated-orcid":false,"given":"Iftikhar","family":"Ahmed","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7367-5920","authenticated-orcid":false,"given":"Anushree","family":"Brahmacharimayum","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9068-2337","authenticated-orcid":false,"given":"Raja Hashim","family":"Ali","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8962-5602","authenticated-orcid":false,"given":"Talha Ali","family":"Khan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7885-0369","authenticated-orcid":false,"given":"Muhammad Ovais","family":"Ahmad","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,9,11]]},"reference":[{"issue":"10312","key":"R1","doi-asserted-by":"publisher","first-page":"1700","DOI":"10.1016\/S0140-6736(21)02143-7","article-title":"Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic","volume":"398","author":"Santomauro","journal-title":"The Lancet"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.i5320","article-title":"Women have higher rates of mental disorders than men, NHS survey finds","author":"Gulland","journal-title":"BMJ"},{"key":"R3","doi-asserted-by":"publisher","DOI":"10.1016\/j.pnpbp.2020.110236","article-title":"Younger people are more vulnerable to stress, anxiety and depression during COVID-19 pandemic: a global cross-sectional survey","volume":"109","author":"Varma","journal-title":"Prog Neuropsychopharmacol Biol Psychiatry"},{"key":"R4","unstructured":"World health organization. Depressive disorder (depression) - who.int. 2023. URL: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/depression [Accessed 04-09-2025]"},{"key":"R5","unstructured":"World health organization. Suicide - India - who.int. URL: https:\/\/www.who.int\/india\/health-topics\/suicide [Accessed 04-09-2025]"},{"key":"R6","unstructured":"Chand SP Arif H . Depression. StatPearls Publishing; 2024. URL: https:\/\/www.ncbi.nlm.nih.gov\/books\/NBK430847\/ [Accessed 04-09-2025]"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2018.00450","article-title":"DSM-5 criteria and depression severity: implications for clinical practice","volume":"9","author":"Tolentino","journal-title":"Front Psychiatry"},{"issue":"6","key":"R8","doi-asserted-by":"publisher","first-page":"289","DOI":"10.9740\/mhc.2016.11.289","article-title":"Medical mimics: differential diagnostic considerations for psychiatric symptoms","volume":"6","author":"McKee","journal-title":"Ment Health Clin"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1146\/annurev-med-061212-171507","article-title":"Depression as a risk factor for cancer: from pathophysiological advances to treatment implications","volume":"65","author":"Currier","journal-title":"Annu Rev Med"},{"issue":"4","key":"R10","doi-asserted-by":"publisher","first-page":"561","DOI":"10.1016\/j.neubiorev.2013.02.004","article-title":"Depression after myocardial infarction: TNF-\u03b1-induced alterations of the blood\u2013brain barrier and its putative therapeutic implications","volume":"37","author":"Liu","journal-title":"Neuroscience & Biobehavioral Reviews"},{"issue":"11","key":"R11","first-page":"858","volume":"82","author":"Kohn","journal-title":"Bull World Health Organ"},{"issue":"6","key":"R12","first-page":"1001","volume":"66","author":"Sharp","journal-title":"Am Fam Physician"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"382","DOI":"10.1192\/bjp.134.4.382","article-title":"A new depression scale designed to be sensitive to change","volume":"134","author":"Montgomery","journal-title":"Br J Psychiatry"},{"issue":"3","key":"R14","doi-asserted-by":"publisher","first-page":"175","DOI":"10.1002\/mpr.1388","article-title":"The structure of the Montgomery\u2013\u00c5sberg depression rating scale over the course of treatment for depression","volume":"22","author":"Quilty","journal-title":"Int J Methods Psych Res"},{"issue":"3","key":"R15","doi-asserted-by":"publisher","first-page":"618","DOI":"10.1161\/01.str.29.3.618","article-title":"The Sunnybrook Stroke Study: a prospective study of depressive symptoms and functional outcome","volume":"29","author":"Herrmann","journal-title":"Stroke"},{"issue":"12","key":"R16","doi-asserted-by":"publisher","DOI":"10.3390\/brainsci11121633","article-title":"Biological, psychological, and social determinants of depression: a review of recent literature","volume":"11","author":"Remes","journal-title":"Brain Sci"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.comppsych.2012.06.006","article-title":"The diagnosis of depression: current and emerging methods","volume":"54","author":"Smith","journal-title":"Compr Psychiatry"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"16381","DOI":"10.1038\/s41598-020-72289-9","article-title":"Using machine learning-based analysis for behavioral differentiation between anxiety and depression","volume":"10","author":"Richter","journal-title":"Sci Rep"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2024.3491593","article-title":"Randomized explainable machine learning models for efficient medical diagnosis","volume":"PP","author":"Muhammad","journal-title":"IEEE J Biomed Health Inform"},{"issue":"1","key":"R20","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1038\/s41591-018-0300-7","article-title":"High-performance medicine: the convergence of human and artificial intelligence","volume":"25","author":"Topol","journal-title":"Nat Med"},{"key":"R21","doi-asserted-by":"publisher","DOI":"10.2196\/42672","article-title":"Wearable artificial intelligence for anxiety and depression: scoping review","volume":"25","author":"Abd-Alrazaq","journal-title":"J Med Internet Res"},{"key":"R22","doi-asserted-by":"crossref","unstructured":"Paul N Juliet S . Comparative analysis of machine learning techniques for mental health prediction. Presented at: 2023 8th International Conference on Communication and Electronics Systems (ICCES); Coimbatore, India. 2023.[doi: 10.1109\/ICCES57224.2023.10192763]","DOI":"10.1109\/ICCES57224.2023.10192763"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-62713-5","article-title":"Classifying post-traumatic stress disorder using the magnetoencephalographic connectome and machine learning","volume":"10","author":"Zhang","journal-title":"Sci Rep"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-17126-x","article-title":"Training a machine learning classifier to identify ADHD based on real-world clinical data from medical records","volume":"12","author":"Mikolas","journal-title":"Sci Rep"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","DOI":"10.1038\/s41398-022-01989-9","article-title":"Patterns of activity correlate with symptom severity in major depressive disorder patients","volume":"12","author":"Spulber","journal-title":"Transl Psychiatry"},{"key":"R26","doi-asserted-by":"crossref","unstructured":"Garcia-Ceja E Riegler M Jakobsen P et al. Motor activity based classification of depression in unipolar and bipolar patients. Presented at: 2018 IEEE 31st International Symposium on Computer-Based Medical Systems (CBMS); Karlstad. 2018.[doi: 10.1109\/CBMS.2018.00062]","DOI":"10.1109\/CBMS.2018.00062"},{"key":"R27","doi-asserted-by":"crossref","unstructured":"Frogner JI Noori FM Halvorsen P et al. One-dimensional convolutional neural networks on motor activity measurements in detection of depression. Presented at: MM \u201919; 9-15; Nice France. Oct 15, 2019.URL: https:\/\/dl.acm.org\/doi\/proceedings\/10.1145\/3347444 [doi: 10.1145\/3347444.3356238]","DOI":"10.1145\/3347444.3356238"},{"issue":"8","key":"R28","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0231995","article-title":"Applying machine learning in motor activity time series of depressed bipolar and unipolar patients compared to healthy controls","volume":"15","author":"Jakobsen","journal-title":"PLoS ONE"},{"issue":"9","key":"R29","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering9090458","article-title":"Two-dimensional convolutional neural network for depression episodes detection in real time using motor activity time series of depresjon dataset","volume":"9","author":"Espino-Salinas","journal-title":"Bioengineering (Basel)"},{"key":"R30","doi-asserted-by":"crossref","unstructured":"Li Y . Application of machine learning to predict mental health disorders and interpret feature importance. Presented at: 2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS). 257-261; 2023.[doi: 10.1109\/ISCTIS58954.2023.10213032]","DOI":"10.1109\/ISCTIS58954.2023.10213032"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.sciaf.2023.e01716","article-title":"Machine learning based model for detecting depression during Covid-19 crisis","volume":"20","author":"Malik","journal-title":"Sci Afr"},{"key":"R32","doi-asserted-by":"publisher","DOI":"10.1016\/j.lanepe.2022.100391","article-title":"Predictors of depression among middle-aged and older men and women in Europe: a machine learning approach","volume":"18","author":"Handing","journal-title":"Lancet Reg Health Eur"},{"issue":"12","key":"R33","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0261131","article-title":"Detection of child depression using machine learning methods","volume":"16","author":"Haque","journal-title":"PLoS ONE"},{"issue":"10","key":"R34","doi-asserted-by":"publisher","first-page":"129","DOI":"10.13053\/rcs-148-10-11","article-title":"Evaluation of five classifiers for depression episodes detection","volume":"148","author":"Pacheco-Gonz\u00e1lez","journal-title":"RCS"},{"issue":"1","key":"R35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1155\/2019\/8269695","article-title":"Depression episodes detection in unipolar and bipolar patients: a methodology with feature extraction and feature selection with genetic algorithms using activity motion signal as information source","volume":"2019","author":"Galv\u00e1n-Tejada","journal-title":"Mobile Information Systems"},{"key":"R36","doi-asserted-by":"crossref","unstructured":"Garcia-Ceja E Riegler M Jakobsen P et al. Depresjon: a motor activity database of depression episodes in unipolar and bipolar patients. Presented at: Multimedia System Conference 2018; Amsterdam. 2018.[doi: 10.1145\/3204949.3208125]","DOI":"10.1145\/3204949.3208125"},{"key":"R37","doi-asserted-by":"crossref","unstructured":"Garcia EA He H Bai Y Li S. A . ADASYN: Adaptive synthetic sampling approach for imbalanced learning. Presented at: 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence); 1322-1328; Hong Kong. 2008.[doi: 10.1109\/IJCNN.2008.4633969]","DOI":"10.1109\/IJCNN.2008.4633969"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics9010008","article-title":"Feature extraction in motor activity signal: towards a depression episodes detection in unipolar and bipolar patients","volume":"9","author":"Zanella-Calzada","journal-title":"Diagnostics (Basel)"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-52905-8","article-title":"General spectral characteristics of human activity and its inherent scale-free fluctuations","volume":"14","author":"Macz\u00e1k","journal-title":"Sci Rep"},{"key":"R40","doi-asserted-by":"crossref","unstructured":"Wilson CE Crouse JJ Carpenter JS Hickie IB . The circadian component of mood disorders: the sleep-wake cycle, biological rhythms, and chronotherapeutics. In: Kushida CA , editor. Encyclopedia of Sleep and Circadian Rhythms. 2nd ed. Academic Press; 2023:766-779. [doi: 10.1016\/B978-0-12-822963-7.00147-X] ISBN: 978-0-323-91094-1","DOI":"10.1016\/B978-0-12-822963-7.00147-X"},{"key":"R41","doi-asserted-by":"crossref","unstructured":"Langensiepen C Lotfi A . Uncertainty measures in an ambient intelligence environment. Presented at: 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). 1-5; 2017.[doi: 10.1109\/FUZZ-IEEE.2017.8015765]","DOI":"10.1109\/FUZZ-IEEE.2017.8015765"},{"key":"R42","doi-asserted-by":"crossref","unstructured":"Hosmer Jr DW Lemeshow S Sturdivant RX . Applied Logistic Regression. John Wiley & Sons; 2013. [doi: 10.1002\/9781118548387]","DOI":"10.1002\/9781118548387"},{"key":"R43","doi-asserted-by":"crossref","unstructured":"Do TN Poulet F . Parallel multiclass logistic regression for classifying large scale image datasets. In: Advanced Computational Methods for Knowledge Engineering Advances in Intelligent Systems and Computing. Springer; 2015:255-266. [doi: 10.1007\/978-3-319-17996-4_23]","DOI":"10.1007\/978-3-319-17996-4_23"},{"issue":"5","key":"R44","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s41870-017-0080-1","article-title":"Survey on SVM and their application in image classification","volume":"13","author":"Chandra","journal-title":"Int j inf tecnol"},{"key":"R45","doi-asserted-by":"crossref","unstructured":"Breiman L . Random forests. In: Machine Learning. Vol 45. Springer; 2001:5-32. [doi: 10.1023\/A:1010933404324]","DOI":"10.1023\/A:1010933404324"},{"key":"R46","doi-asserted-by":"crossref","unstructured":"Cutler A Cutler DR Stevens JR . Random forests. In: Ensemble Machine Learning: Methods and Applications. Springer; 2012:157-175. [doi: 10.1007\/978-1-4419-9326-7_5]","DOI":"10.1007\/978-1-4419-9326-7_5"},{"key":"R47","doi-asserted-by":"crossref","unstructured":"Chen T Guestrin C . Xgboost: a scalable tree boosting system. arXiv. Preprint posted online on Mar 19, 2016. [doi: 10.48550\/arXiv.1603.02754]","DOI":"10.1145\/2939672.2939785"},{"key":"R48","doi-asserted-by":"crossref","unstructured":"Rahman RU Ahmad I Li Z Sparks R Saad AB Mullen A . Singletrome: a method to analyze and enhance the transcriptome with long noncoding rnas for single cell analysis. bioRxiv. Nov 2, 2022. [doi: 10.1101\/2022.10.31.514182v1]","DOI":"10.1101\/2022.10.31.514182"},{"key":"R49","unstructured":"Goodfellow I Bengio Y Courville A . Deep Learning. The MIT Press; 2016. ISBN: 9780262035613"},{"key":"R50","doi-asserted-by":"crossref","unstructured":"Misgar MM Bhatia MPS . Detection of depression from iomt time series data using UMAP features. Presented at: 2022 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS). 623-628; 2022.[doi: 10.1109\/ICCCIS56430.2022.10037751]","DOI":"10.1109\/ICCCIS56430.2022.10037751"},{"issue":"4","key":"R51","doi-asserted-by":"publisher","first-page":"1335","DOI":"10.32604\/iasc.2020.013861","article-title":"Experimental evaluation of clickbait detection using machine learning models","volume":"26","author":"Ahmad","journal-title":"Intelligent Automation & Soft Computing"},{"key":"R52","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2024.08.005","article-title":"Unveiling the black box: a systematic review of explainable artificial intelligence in medical image analysis","volume":"24","author":"Muhammad","journal-title":"Comput Struct Biotechnol J"},{"key":"R53","unstructured":"Su-In L Scott M . A unified approach to interpreting model predictions. arXiv. Preprint posted online on May 22, 2017. [doi: 10.48550\/arXiv.1705.07874]"},{"key":"R54","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2023.1214143","article-title":"Rest-activity rhythm associated with depressive symptom severity and attention among patients with major depressive disorder: a 12-month follow-up study","volume":"14","author":"Yang","journal-title":"Front Psychiatry"},{"issue":"1","key":"R55","doi-asserted-by":"publisher","DOI":"10.1038\/s41398-020-0694-0","article-title":"Circadian rhythm disruption and mental health","volume":"10","author":"Walker","journal-title":"Transl Psychiatry"},{"key":"R56","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2024.105094","article-title":"Causal dynamics of sleep, circadian rhythm, and mood symptoms in patients with major depression and bipolar disorder: insights from longitudinal wearable device data","volume":"103","author":"Song","journal-title":"EBioMedicine"},{"issue":"11","key":"R57","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics15111412","article-title":"Explainable machine learning in the prediction of depression","volume":"15","author":"Mimikou","journal-title":"Diagnostics (Basel)"},{"issue":"8","key":"R58","doi-asserted-by":"publisher","first-page":"783","DOI":"10.1037\/bul0000102","article-title":"Gender differences in depression in representative national samples: meta-analyses of diagnoses and symptoms","volume":"143","author":"Salk","journal-title":"Psychol Bull"},{"issue":"4","key":"R59","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1016\/j.jagp.2012.09.003","article-title":"Depressive symptoms and circadian activity rhythm disturbances in community-dwelling older women","volume":"22","author":"Maglione","journal-title":"Am J Geriatr Psychiatry"},{"issue":"2","key":"R60","doi-asserted-by":"publisher","first-page":"114","DOI":"10.1159\/000441458","article-title":"Critical slowing down as a personalized early warning signal for depression","volume":"85","author":"Wichers","journal-title":"Psychother Psychosom"},{"issue":"5","key":"R61","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1379","article-title":"Interpretability of machine learning\u2010based prediction models in healthcare","volume":"10","author":"Stiglic","journal-title":"WIREs Data Min & Knowl"},{"key":"R62","unstructured":"Simula. The Depresjon Dataset. URL: https:\/\/datasets.simula.no\/depresjon\/ [Accessed 04-09-2025]"}],"container-title":["JMIR Mental Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/mental.jmir.org\/2025\/1\/e72038","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:02:07Z","timestamp":1782842527000},"score":1,"resource":{"primary":{"URL":"https:\/\/mental.jmir.org\/2025\/1\/e72038"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,11]]},"references-count":62,"URL":"https:\/\/doi.org\/10.2196\/72038","relation":{},"ISSN":["2368-7959"],"issn-type":[{"value":"2368-7959","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,11]]},"article-number":"v12i10312e72038"}}