Quantitative prediction of radiographic progression in patients with axial spondyloarthritis using neural network model in a real-world setting

• Published in: Arthritis research & therapy Vol. 25; no. 1; p. 65
• Main Authors: Baek, In-Woon, Jung, Seung Min, Park, Yune-Jung, Park, Kyung-Su, Kim, Ki-Jo
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 20.04.2023; BioMed Central; BMC
• Subjects: Algorithms; Analysis; Ankylosing spondylitis; Arthritis; Artificial neural network; Axial spondyloarthritis; Back propagation; Care and treatment; Datasets; Development and progression; Disease Progression; Generalized linear models; Health aspects; Humans; Machine learning; Medical records; Methods; Neural networks; Patients; Performance evaluation; Quality of life; Quantitative prediction; Radiographic progression; Radiography; Real-world setting; Sacroiliitis; Severity of Illness Index; Spine; Spondylarthritis - diagnostic imaging; Spondylitis, Ankylosing - diagnostic imaging; TNF inhibitors; Tumor necrosis factor-TNF; Variables; South Korea; Real-world setting; Axial spondyloarthritis; Artificial neural network; Quantitative prediction; Radiographic progression
• ISSN: 1478-6362; 1478-6354; 1478-6362
• DOI: 10.1186/s13075-023-03050-6

Predicting radiographic progression in axial spondyloarthritis (axSpA) remains limited because of the complex interaction between multiple associated factors and individual variability in real-world settings. Hence, we tested the feasibility of artificial neural network (ANN) models to predict radiographic progression in axSpA. In total, 555 patients with axSpA were split into training and testing datasets at a 3:1 ratio. A generalized linear model (GLM) and ANN models were fitted based on the baseline clinical characteristics and treatment-dependent variables for the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) of the radiographs at follow-up time points. The mSASSS prediction was evaluated, and explainable machine learning methods were used to provide insights into the model outcome or prediction. The R values of the fitted models were in the range of 0.90-0.95 and ANN with an input of mSASSS as the number of each score performed better (root mean squared error (RMSE) = 2.83) than GLM or input of mSASSS as a total score (RMSE = 2.99-3.57). The ANN also effectively captured complex interactions among variables and their contributions to the transition of mSASSS over time in the fitted models. Structural changes constituting the mSASSS scoring systems were the most important contributing factors, and no detectable structural abnormalities at baseline were the most significant factors suppressing mSASSS change. Clinical and radiographic data-driven ANN allows precise mSASSS prediction in real-world settings. Correct evaluation and prediction of spinal structural changes could be beneficial for monitoring patients with axSpA and developing a treatment plan.