Deep-learning based quantification model for hip bone marrow edema and synovitis in patients with spondyloarthritis based on magnetic resonance images

Abstract

Objectives: Hip inflammation is one of the most common complications in patients with spondyloarthritis (SpA). Herein, we employed use of a deep learning-based magnetic resonance imaging (MRI) evaluation model to identify irregular and multiple inflammatory lesions of the hip.Methods: All of the SpA patients were enrolled at the Xijing Hospital. The erythrocyte sediment rate (ESR), C-reactive protein (CRP), hip function Harris score, and disease activity were evaluated by clinicians. Manual MRI annotations including bone marrow edema (BME) and effusion/synovitis, and a hip MRI scoring system (HIMRISS) assessment was performed by experienced musculoskeletal radiologists. The segmentation accuracies of four deep learning models, including U-Net, UNet++, Attention-Unet, and HRNet, were compared using five-fold cross-validation. The clinical agreement of U-Net was evaluated with clinical symptoms and HIMRISS results.Results: A total of 1945 MRI slices of STIR/T2WI sequences were obtained from 195 SpA patients with hip involvement. After the five-fold cross-validation, U-Net achieved an average segmentation accuracy of 88.48% for the femoral head and 69.36% for inflammatory lesions, which are higher than those obtained by the other three models. The UNet-score, which was calculated based on the same MRI slices as HIMRISS, was significantly correlated with the HIMRISS scores and disease activity indexes (p values <0.05).Conclusion: This deep-learning based automatic MRI evaluation model could achieve similar quantification performance as an expert radiologist, and it has the potential to improve the accuracy and efficiency of clinical diagnosis for SpA patients with hip involvement.

Implementation of artificial intelligence models in magnetic resonance imaging with focus on diagnosis of rheumatoid arthritis and axial spondyloarthritis: narrative review

“…Zheng et al ( 76 ) utilized a DL-driven MRI image assessment technique to detect inflammatory alterations in the hip among individuals with axSpA. The findings closely matched those of specialized radiologists, underscoring the substantial potential of this model to enhance diagnostic precision in axSpA patients, particularly emphasizing its relevance in clinical evaluations.…”

ASNET: A Novel AI Framework for Accurate Ankylosing Spondylitis Diagnosis from MRI

“…The average accuracy, sensitivity, and specificity values were 91.60%, 80.28%, and 94.24%, respectively. Zheng et al [ 38 ] presented a deep learning-based model for hip bone marrow edema and synovitis in spondyloarthritis patients using MRI. They compared four deep learning models and found that U-Net achieved segmentation accuracy for femoral heads and inflammatory lesions.…”

Artificial intelligence and machine learning in axial spondyloarthritis

“…Bordner et al [37] and Bressem et al [38,39 && ] illustrate how artificial intelligence models accurately identify structural and inflammatory changes according to ASAS criteria, yielding comparable performance to radiologists and highlighting the importance of early detection for positive patient outcomes. Lin et al [40] successfully applied deep learning in line with the SPARCC scoring system to grade sacroiliitis on MRI, showing significant alignment with expert evaluations, and other researcher employed deep learning based models to identify, sometimes quantify BME and synovitis in axSpA patients [41][42][43]. High accuracy in BME is critical in axSpA, as early detection of inflammation alters the course of the disease.…”