Artificial intelligence can help in the diagnosis of axSPA

In a study, an artificial neural network was remarkably successful in the assessment of radiographic sacroiliitis.

The reliability of radiographic sacroiliitis assessment is known to be poor. Expert readers –as opposed to evaluation by local rheumatologists or radiologists– usually produce more reliable results, but they are not available in many locations. “We see a big discrepancy between the local and central assessment of sacroiliitis reaching sometimes half of the cases,” Prof. Denis Poddubnyy (Charité Universitätsmedizin Berlin, Germany) said. Can artificial intelligence support the diagnosis?

For this study, Prof. Poddubnyy and his team used conventional radiographs of sacroiliac joints from 2 independent cohorts of patients with axial spondylarthritis (axSpA), including 1,669 radiographs used to train and validate the neural network, and 100 radiographs used as a test dataset [1]. All radiographs went through reading by both humans and the artificial neural network. Readers used the modified New York criteria to determine either the presence or absence of definite radiographic sacroiliitis. The researchers then analysed whether the human readers or artificial neural network agreed.

The artificial neural network achieved excellent performance in accurately recognising definite radiographic sacroiliitis in these patients, with high ratings of sensitivity and specificity (0.90 and 0.93 for the validation and 0.87 and 0.97 for the test set). This artificial intelligence-driven model could enable accurate detection of sacroiliitis for both diagnosis of patients in the clinic and classification of axSpA when selecting patients for clinical trials.

“I do think that the developed artificial neural network might be helpful in clinical practice,” Prof. Poddubnyy concluded. This approach will now be tested for the assessment of MRI of sacroiliac joints, which would be especially relevant for the diagnosis of axSpA in an early stage.

1. Bressem KK, et al. Development and validation of an artificial intelligence approach for the detection of radiographic sacroiliitis. 2018, ACR Convergence 2020, 5-9 Nov.

Posted on 18 January 202117 May 2024