medwireNews: Machine learning models have demonstrated potential for predicting radiographic progression in patients with axial spondyloarthritis (axSpA).
Ki-Jo Kim and colleagues from The Catholic University of Korea in Seoul used data from a training cohort of 253 patients to develop seven predictive models with algorithms based on clinical and laboratory data, and tested the models in a cohort of 173 patients. In all, 25.3% and 23.7% of patients in the training and testing cohorts, respectively, experienced radiographic progression over 2 years.
The researchers report in Clinical Rheumatology that predicting progression was “achievable with promising results,” with the most important predictors including mSASSS, baseline syndesmophytes, and sacroiliac joint grades.
In area under the receiver operating characteristic curve analysis, the two best-performing models – generalized linear model and support vector machine – correctly distinguished between patients with and without radiographic progression in the testing cohort on an average of 78% of occasions.
These findings indicate that the models have “reasonable performance in the prediction of radiographic progression in axSpA,” write Kim and team.
“Further modelling with larger and more detailed data could provide an excellent opportunity for the clinical translation of the predictive models to the management of high-risk patients,” they conclude.
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group