Interpretable machine learning integrates multi-source biomarkers for osteoarthritis diagnosis and mechanistic insights: A temporomandibular joint model

Najla Al Turkestani; Lucia Cevidanes; Jonas Bianchi; James Sugai; Marcela Gurgel; Juan Prieto; Elizabeth Hatfield; Kristine Philips; Erika Benavides; Fabiana Soki; Yuji Mishina; Margherita Fontana; Arvind Rao; Hongtu Zhu; Tengfei Li

doi:10.1016/j.joca.2025.08.002

Interpretable machine learning integrates multi-source biomarkers for osteoarthritis diagnosis and mechanistic insights: A temporomandibular joint model

Osteoarthritis Cartilage. 2025 Aug 14:S1063-4584(25)01112-4. doi: 10.1016/j.joca.2025.08.002. Online ahead of print.

Authors

Najla Al Turkestani¹, Lucia Cevidanes², Jonas Bianchi³, James Sugai⁴, Marcela Gurgel⁵, Juan Prieto⁶, Elizabeth Hatfield⁷, Kristine Philips⁸, Erika Benavides⁹, Fabiana Soki¹⁰, Yuji Mishina¹¹, Margherita Fontana¹², Arvind Rao¹³, Hongtu Zhu¹⁴, Tengfei Li¹⁵

Affiliations

¹ Department of Restorative Dentistry, Faculty of Dentistry, King Abdulaziz University, Jeddah, Makkah Province 21589, Saudi Arabia; Department of Orthodontics and Pediatric Dentistry, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: alturkistani@kau.edu.sa.
² Department of Orthodontics and Pediatric Dentistry, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: luciacev@umich.edu.
³ Department of Orthodontics, University of the Pacific, Arthur A. Dugoni School of Dentistry, San Francisco, CA 94103, United States. Electronic address: jbianchi@pacific.edu.
⁴ Department of Periodontics & Oral Medicine, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: jsugai@umich.edu.
⁵ Department of Orthodontics and Pediatric Dentistry, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: mlimagur@umich.edu.
⁶ Department of Psychiatry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, United States. Electronic address: juanprietob@gmail.com.
⁷ Department of Oral and Maxillofacial Surgery, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: lizhat@umich.edu.
⁸ Department of Oral and Maxillofacial Surgery, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: phillikr@umich.edu.
⁹ Department of Periodontics & Oral Medicine, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: benavid@umich.edu.
¹⁰ Department of Periodontics & Oral Medicine, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: fabisoki@umich.edu.
¹¹ Department of Biologic and Materials Sciences & Prosthodontics, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: mishina@umich.edu.
¹² Department of Cariology, Restorative Sciences and Endodontics, School of Dentistry, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: mfontan@umich.edu.
¹³ Department of Radiation Oncology, 2Department of Computational Medicine & Bioinformatics, University of Michigan, Ann Arbor, MI 48109, United States. Electronic address: ukarvind@umich.edu.
¹⁴ Department of Biostatistics and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, United States. Electronic address: htzhu@email.unc.edu.
¹⁵ Department of Radiology and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, United States. Electronic address: tengfei_li@med.unc.edu.

PMID: 40818794
PMCID: PMC12452200 (available on 2026-08-14)
DOI: 10.1016/j.joca.2025.08.002

Abstract

Objective: Osteoarthritis (OA) is a pan-joint degenerative disorder characterized by cartilage degradation and subchondral bone remodeling. The temporomandibular joint (TMJ) offers a unique model for early OA due to its anatomy and early-onset disease. Current diagnostics rely on late-stage changes, underscoring the need for biomarker integration. We hypothesized that machine learning (ML) combining imaging, molecular, and clinical data would improve diagnostic accuracy, and that SHapley Additive exPlanations (SHAP) would clarify key predictors and interactions, enhancing mechanistic understanding of disease heterogeneity.

Design: A case-control study of 162 participants (81 TMJ OA and 81 age- and sex-matched controls) integrated clinical, high-resolution imaging (radiomics, trabecular architecture, joint space), and systemic/articular biomarkers (serum and saliva). Seventy-seven ML combinations were evaluated via nested 10-fold cross-validation.

Results: The final ensemble model achieved strong diagnostic performance (AUC=0.828, 95% CI: 0.757-0.892). SHAP analysis revealed top predictors such as headache severity, trabecular thickening, restless sleep, muscle soreness, limited mouth opening and joint space narrowing. Mechanistic interactions captured early inflammatory, structural, and neurovascular changes, including radiomics-cartilage degradation links (e.g., condyle grey level nonuniformity with saliva CXCL-16), clinical-molecular associations (e.g., headaches with saliva VE-cadherin), and subchondral microstructure correlations (e.g., grey level nonuniformity with run length nonuniformity).

Conclusions: This study presents a clinically useful, explainable AI model for OA diagnosis. Key predictors and cross-domain interactions improved accuracy and clarified early disease mechanisms. Although cross-validation minimized overfitting risk, external validation is needed. These findings support biomarker-driven precision diagnostics and highlight multi-tissue predictors as potential targets for early OA intervention.

Keywords: Cartilage degeneration; Diagnostic biomarkers; Machine learning; Osteoarthritis; Subchondral bone remodeling.

Grants and funding

R01 DE024450/DE/NIDCR NIH HHS/United States