Detecting radiographic sacroiliitis using deep learning with expert-level accuracy in axial spondyloarthritis

Objectives To develop and validate an artificial neural network for the detection of definite radiographic sacroiliitis as a manifestation of axial spondyloarthritis. Methods Conventional radiographs of sacroiliac joints from two independent cohorts of patients with axial spondyloarthritis (axSpA) were used. The first cohort consisted of 1669 radiographs and was used for training and validation of a neural network. The second cohort consisted of 525 radiographs, of which 100 radiographs were randomly selected for the test dataset. In both cohorts all radiographs underwent central reading; the final decision on the presence or absence of definite radiographic sacroiliitis was used as a reference. For performance evaluation of the neural network, areas under the receiver operating characteristic curves (AUROC) were calculated. Sensitivity and specificity for the prediction cut-offs were calculated. Cohen’s Kappa and the absolute agreement were used to assess the agreement between the neural network and the human readers. Results The neural network achieved an excellent performance in recognition of definite radiographic sacroiliitis with AUROC of 0.97 and 0.96 for the validation and test datasets, respectively. Sensitivity and specificity for the cut-off weighting both measurements equally were 0.90 and 0.93 for the validation and 0.87 and 0.97 for the test set. The Cohen’s kappa between the neural network and the reference judgements were 0.80 for both validation and test sets, and the absolute agreement on the classification yielded 91% and 90%, respectively. Conclusions Artificial neural networks enable the accurate detection of definite radiographic sacroiliitis relevant for the diagnosis and classification of axSpA.