Advancing osteoarthritis research: the role of AI in clinical, imaging and omics fields

Abstract

Osteoarthritis (OA) is a degenerative joint disease with significant clinical and societal impact. Traditional diagnostic methods, including subjective clinical assessments and imaging techniques such as X-rays and MRIs, are often limited in their ability to detect early-stage OA or capture subtle joint changes. These limitations result in delayed diagnoses and inconsistent outcomes. Additionally, the analysis of omics data is challenged by the complexity and high dimensionality of biological datasets, making it difficult to identify key molecular mechanisms and biomarkers. Recent advancements in artificial intelligence (AI) offer transformative potential to address these challenges. This review systematically explores the integration of AI into OA research, focusing on applications such as AI-driven early screening and risk prediction from electronic health records (EHR), automated grading and morphological analysis of imaging data, and biomarker discovery through multi-omics integration. By consolidating progress across clinical, imaging, and omics domains, this review provides a comprehensive perspective on how AI is reshaping OA research. The findings have the potential to drive innovations in personalized medicine and targeted interventions, addressing longstanding challenges in OA diagnosis and management.

Fig. 1

Workflow for clinical data analysis and AI-driven applications in OA: the process begins with data acquisition from various clinical sources, including basic demographic data, clinical visit records, laboratory test results, and electronic medical records (a). These data are analyzed using machine learning (e.g., Random Forest, SVM, XGBoost, K-NN) and deep learning methods (e.g., deep neural networks, convolutional neural networks, U-Net) to identify latent pathological features and risk factors associated with OA (b). The resulting insights are applied to a range of clinical tasks, such as OA risk prediction, surgery prediction, recovery prediction, and other applications, facilitating early screening, personalized treatment strategies, and improved patient outcomes (c)

Fig. 2

Workflow of AI-enhanced diagnosis and treatment of osteoarthritis: from image acquisition to rapid clinical decision support. AI first performs preprocessing of the raw imaging data. The algorithm automatically extracts key information from the joint structures, such as morphological changes in cartilage and bone, and identifies potential lesion areas. Deep learning models are used to detect abnormal structures within the joint at the early stages of OA. Finally, AI technology integrates clinical and omics data to provide a more comprehensive assessment of OA progression. By enhancing the disease diagnosis and treatment workflow with AI, the sensitivity of early OA screening can be significantly improved, and personalized treatment plans can be developed

Fig. 3

Traditional workflow for OA image data analysis and diagnosis. a Acquire imaging data (e.g., X-ray, MRI, CT, ultrasound) from clinical sources. b Preprocess images (resizing, denoising, data augmentation) and perform segmentation. c Extract relevant features (e.g., texture, shape). d Statistical analysis and integrate results into clinical decision support

Fig. 4

Workflow for OA transcriptomic biomarker discovery. a Obtain scRNA-seq and bulk RNA-seq datasets from databases like GEO. b Identify differentially expressed genes using analytical tools (e.g., limma, edgeR, DESeq2). c Perform functional and pathway enrichment analysis (e.g., KEGG, GO, DO, PPI). d Identify genes associated with OA diagnosis by machine learning models (e.g., Random Forest, SVM, XGBoost). e Validate biomarkers in other datasets and investigate potential treatments