Assessment of a Machine Learning Tool for Early Identification of Psoriatic Arthritis

According to a study published in the Journal of Translational Autoimmunity, a machine learning tool could aid in the early identification of undiagnosed patients with psoriatic arthritis (PsA).

Researchers aimed to assess the efficacy of a proprietary machine learning tool called PredictAI in identifying undiagnosed patients with PsA 1 to 4 years before they were initially suspected of having PsA. Analyzing data from Maccabi Healthcare Service between 2008 and 2020, the research involved 2 distinct cohorts: the general adult population and the psoriasis cohort. Each cohort was divided into training and testing sets, and the PredictAI model was developed and evaluated based on data from 3 years preceding the reference event.

The results indicated that the machine learning tool successfully identified undiagnosed patients with PsA in the psoriasis cohort with a specificity of 90% 1 to 4 years before the reference event. The sensitivity was 51% and 38%, and the positive predictive value (PPV) stood at 36.1% and 29.6%, respectively. In the general population cohort, the model achieved a sensitivity of 43% and 32% and a PPV of 10.6% and 8.1% for the same time windows, with a specificity of 99%.

“The presented machine learning tool may aid in the early identification of undiagnosed PsA patients, and thereby promote earlier intervention and improve patient outcomes,” the authors concluded.

Reference
Shapiro J, Getz B, Cohen SB, et al. Evaluation of a machine learning tool for the early identification of patients with undiagnosed psoriatic arthritis—a retrospective population-based study. J Transl Autoimmun. 2023;7:100207. Published online August 2, 2023.  doi:10.1016/j.jtauto.2023.100207