{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T15:16:36Z","timestamp":1783523796046,"version":"3.55.0"},"reference-count":28,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"unspecified","delay-in-days":7,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Health Care Inform"],"accepted":{"date-parts":[[2026,6,21]]},"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>To develop a robust and interpretable machine learning framework for arthritis risk prediction and to identify important risk factors associated with the disease.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study used four datasets, which are Behavioral Risk Factor Surveillance System (BRFSS) 2019, BRFSS 2021, National Health Interview Survey (NHIS) 2020 and NHIS 2022. We evaluated 11 machine learning and deep learning architectures including a custom stacked ensemble combined with five resampling techniques to address class imbalance and 9 imputation methods to handle missing data. Area under the receiver operating characteristic curve (AUROC), sensitivity, specificity and balanced accuracy are used to evaluate model performance, with fivefold cross-validation to ensure robustness. SHapley Additive exPlanations (SHAP) provided feature interpretability in prediction.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The combination of the stacked ensemble, Generative Adversarial Network (GAN) imputation and random undersampling (RUS) achieved the best performance with an AUROC of 0.8007 and sensitivity of 0.8060 in the test set of BRFSS 2019. This combination also showed a balanced performance across both subgroups of male and female by maintaining high performance with AUROC of 0.808 and 0.800, respectively, in the test set of BRFSS 2021. The stacked ensemble model with GAN imputation and either Random Undersampling (RUS) or Random Oversampling (ROS) achieved the best performance across all four datasets. SHAP analysis identified higher age, walking difficulty and high body mass index (BMI) as the top physical factors. Among the psychosocial factors, parental separation showed a high impact on predicting arthritis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The risk of arthritis is influenced by health, lifestyle and childhood experiences. SHAP shows that difficulty walking, age, BMI and childhood stress play important roles in predicting arthritis risk.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>These results confirm important predictors of arthritis risk and provide a clear, interpretable framework.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjhci-2025-101955","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T14:35:16Z","timestamp":1783521316000},"page":"e101955","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Explainable machine learning revealing the impact of mental and physical health on arthritis"],"prefix":"10.1136","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-9984-058X","authenticated-orcid":false,"given":"Md. Atik","family":"Shams","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, Notre Dame University Bangladesh, Dhaka, Bangladesh"},{"name":"Department of Computer Science and Engineering, University of Asia Pacific, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5662-314X","authenticated-orcid":false,"given":"Sumaiya","family":"Fatema","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University of Asia Pacific, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-0135-3169","authenticated-orcid":false,"given":"D M","family":"Hasibul Islam","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University of Asia Pacific, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anindita","family":"Datta","sequence":"additional","affiliation":[{"name":"Department of Fish and Wildlife Conservation, Virginia Tech, Blacksburg, Virginia, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Eisenberg","sequence":"additional","affiliation":[{"name":"Department of Information Management and Business Analytics, Montclair State University, Montclair, New Jersey, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Danastan","family":"Tasaouf Mridula","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Northern University Bangladesh, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junnatul","family":"Mawa","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University of Asia Pacific, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5306-5505","authenticated-orcid":false,"given":"Asma","family":"Sultana","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, University of Asia Pacific, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nafiya","family":"Ahmed","sequence":"additional","affiliation":[{"name":"Department of Computer Science, BRAC University, Dhaka, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Monowarul","family":"Islam","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronic Engineering, Islamic University, Kushtia District, Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7325-9844","authenticated-orcid":false,"given":"Tanmoy Sarkar","family":"Pias","sequence":"additional","affiliation":[{"name":"Stanford Medicine, Stanford University, Stanford, California, USA"},{"name":"Department of Computer Science, Virginia Tech, Blacksburg, Virginia, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,7,8]]},"reference":[{"key":"2026070807350969000_33.1.e101955.1","doi-asserted-by":"crossref","first-page":"36","DOI":"10.32481\/djph.2017.03.006","article-title":"Arthritis and its Public Health Burden","volume":"3","author":"Eakin","year":"2017","journal-title":"Dela J Public Health"},{"key":"2026070807350969000_33.1.e101955.2","unstructured":"Centers for Disease Control and Prevention . Arthritis basics. 2025. Available: https:\/\/www.cdc.gov\/arthritis\/basics\/index.html [Accessed 20 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.3","unstructured":"Johns Hopkins Medicine . Arthritis. 2025. Available: https:\/\/www.hopkinsmedicine.org\/health\/conditions-and-diseases\/arthritis [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.4","unstructured":"Johns Hopkins Medicine . Arthritis Diagnosis, 2021. Available: https:\/\/www.hopkinsmedicine.org\/health\/treatment-tests-and-therapies\/arthritis-diagnosis [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.5","unstructured":"Cleveland Clinic . Arthritis. 2025. Available: https:\/\/my.clevelandclinic.org\/health\/diseases\/12061-arthritis [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.6","doi-asserted-by":"crossref","DOI":"10.1186\/s12911-025-03333-9","article-title":"Explainable extratreeclassifier model for early detection of type 2 diabetes: evidence from the PERSIAN Dena Cohort","volume":"26","author":"Ghaderzadeh","year":"2025","journal-title":"BMC Med Inform Decis Mak"},{"key":"2026070807350969000_33.1.e101955.7","doi-asserted-by":"crossref","DOI":"10.1186\/s12889-025-24953-w","article-title":"Leveraging XGBoost and explainable AI for accurate prediction of type 2 diabetes","volume":"25","author":"Rafie","year":"2025","journal-title":"BMC Public Health"},{"key":"2026070807350969000_33.1.e101955.8","doi-asserted-by":"crossref","DOI":"10.1186\/s12911-025-03255-6","article-title":"Artificial intelligence in polycystic ovary syndrome: a systematic review of diagnostic and predictive applications","volume":"25","author":"Ghaderzadeh","year":"2025","journal-title":"BMC Med Inform Decis Mak"},{"key":"2026070807350969000_33.1.e101955.9","doi-asserted-by":"crossref","unstructured":"Luo X , Yang J , Weng W , et al . Prediction of Disease Activity in Rheumatoid Arthritis Based on Class Imbalance. J Shanghai Jiaotong Univ (Sci) 2025;1\u201311. doi:10.1007\/s12204-025-2840-6","DOI":"10.1007\/s12204-025-2840-6"},{"key":"2026070807350969000_33.1.e101955.10","doi-asserted-by":"crossref","first-page":"1282","DOI":"10.1111\/1756-185X.14203","article-title":"Handling missing data in a rheumatoid arthritis registry using random forest approach","volume":"24","author":"Alsaber","year":"2021","journal-title":"Int J of Rheum Dis"},{"key":"2026070807350969000_33.1.e101955.11","doi-asserted-by":"crossref","DOI":"10.3390\/jcm14113829","article-title":"Missing Data in Orthopaedic Clinical Outcomes Research: A Sensitivity Analysis of Imputation Techniques Utilizing a Large Multicenter Total Shoulder Arthroplasty Database","volume":"14","author":"Hao","year":"2025","journal-title":"J Clin Med"},{"key":"2026070807350969000_33.1.e101955.12","first-page":"1","article-title":"Progressive feature reduction with varied missing data and feature selection for arthritis disease prediction","volume":"48","author":"Ramasamy","year":"2025","journal-title":"IFS"},{"key":"2026070807350969000_33.1.e101955.13","doi-asserted-by":"crossref","first-page":"14326","DOI":"10.1038\/s41598-025-99411-z","article-title":"Constructing machine learning-based risk prediction model for osteoarthritis in population aged 45 and above: NHANES 2011\u20132018","volume":"15","author":"Fu","year":"2025","journal-title":"Sci Rep"},{"key":"2026070807350969000_33.1.e101955.14","doi-asserted-by":"crossref","unstructured":"Khalil K , Niloy SI , Mohaidat T , et al . Efficient deep learning approach for arthritis prediction. 2024 IEEE 2nd International Conference on Artificial Intelligence, Blockchain, and Internet of Things (AIBThings); Mt Pleasant, MI, USA, 2024:1\u20135. doi:10.1109\/AIBThings63359.2024.10863076","DOI":"10.1109\/AIBThings63359.2024.10863076"},{"key":"2026070807350969000_33.1.e101955.15","doi-asserted-by":"crossref","first-page":"109413","DOI":"10.1016\/j.compbiomed.2024.109413","article-title":"Fair and explainable Myocardial Infarction (MI) prediction: Novel strategies for feature selection and class imbalance correction","volume":"184","author":"Akter","year":"2025","journal-title":"Comput Biol Med"},{"key":"2026070807350969000_33.1.e101955.16","doi-asserted-by":"crossref","DOI":"10.1097\/MD.0000000000041892","article-title":"Trends and disparities in the prevalence of diagnosed arthritis among United States adults from 2019 to 2022","volume":"104","author":"Rahman","year":"2025","journal-title":"Medicine (Baltimore)"},{"key":"2026070807350969000_33.1.e101955.17","unstructured":"Centers for Disease Control and Prevention . 2021 BRFSS survey data and documentation. 2021. Available: https:\/\/www.cdc.gov\/brfss\/annual_data\/annual_2021.html [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.18","unstructured":"Centers for Disease Control and Prevention . 2019 BRFSS survey data and documentation. 2019. Available: https:\/\/www.cdc.gov\/brfss\/annual_data\/annual_2019.html [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.19","unstructured":"Centers for Disease Control and Prevention . National health interview survey. Available: https:\/\/www.cdc.gov\/nchs\/nhis\/index.htm [Accessed 21 Oct 2025]."},{"key":"2026070807350969000_33.1.e101955.20","unstructured":"Yoon J , Jordon J , Schaar M . GAIN: missing data imputation using generative adversarial nets. International Conference on Machine Learning; 2018:5689\u201398. doi:10.48550\/arXiv.1806.02920"},{"key":"2026070807350969000_33.1.e101955.21","unstructured":"Xu L , Skoularidou M , Cuesta-Infante A , et al . Modeling tabular data using conditional gan. Advances in Neural Information Processing Systems; 2019 doi:10.48550\/arXiv.1907.00503"},{"key":"2026070807350969000_33.1.e101955.22","unstructured":"Choi E , Biswal S , Malin B , et al . Generating multi-label discrete patient records using generative adversarial networks. Machine Learning for Healthcare Conference; 2017:286\u2013305. doi:10.48550\/arXiv.1703.06490"},{"key":"2026070807350969000_33.1.e101955.23","doi-asserted-by":"crossref","first-page":"409","DOI":"10.1016\/j.humov.2018.11.002","article-title":"Self-reported walking difficulty and knee osteoarthritis influences limb dynamics and muscle co-contraction during gait","volume":"64","author":"Na","year":"2019","journal-title":"Hum Mov Sci"},{"key":"2026070807350969000_33.1.e101955.24","doi-asserted-by":"crossref","first-page":"1028","DOI":"10.1002\/acr.21655","article-title":"Association of slower walking speed with incident knee osteoarthritis\u2013related outcomes","volume":"64","author":"Purser","year":"2012","journal-title":"Arthritis Care & Research"},{"key":"2026070807350969000_33.1.e101955.25","doi-asserted-by":"crossref","DOI":"10.1016\/j.pmedr.2021.101463","article-title":"Association of childhood physical and sexual abuse with arthritis in adulthood: Findings from a population-based study","volume":"23","author":"Baiden","year":"2021","journal-title":"Prev Med Rep"},{"key":"2026070807350969000_33.1.e101955.26","doi-asserted-by":"publisher","DOI":"10.1016\/j.pain.2009.01.034"},{"key":"2026070807350969000_33.1.e101955.27","doi-asserted-by":"crossref","first-page":"277","DOI":"10.1007\/s10067-024-07244-4","article-title":"Risk prediction model for psoriatic arthritis: NHANES data and multi-algorithm approach","volume":"44","author":"Zhan","year":"2025","journal-title":"Clin Rheumatol"},{"key":"2026070807350969000_33.1.e101955.28","doi-asserted-by":"crossref","first-page":"1511","DOI":"10.2147\/JIR.S487595","article-title":"Development and Validation of a Cost-Effective Machine Learning Model for Screening Potential Rheumatoid Arthritis in Primary Healthcare Clinics","volume":"18","author":"Wu","year":"2025","journal-title":"J Inflamm Res"}],"container-title":["BMJ Health &amp; Care Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjhci-2025-101955","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T14:35:19Z","timestamp":1783521319000},"score":1,"resource":{"primary":{"URL":"https:\/\/informatics.bmj.com\/lookup\/doi\/10.1136\/bmjhci-2025-101955"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":28,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,7,8]]},"published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1136\/bmjhci-2025-101955"],"URL":"https:\/\/doi.org\/10.1136\/bmjhci-2025-101955","relation":{},"ISSN":["2632-1009"],"issn-type":[{"value":"2632-1009","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]}}}