Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Aug 2:7:100207.
doi: 10.1016/j.jtauto.2023.100207. eCollection 2023 Dec.

Evaluation of a machine learning tool for the early identification of patients with undiagnosed psoriatic arthritis - A retrospective population-based study

J Shapiro  1 , B Getz  2 , S B Cohen  3 , Y Jenudi  2 , D Underberger  2 , M Dreyfuss  2 , T I Ber  2 , S Steinberg-Koch  2 , A Ben-Tov  4   5 , Y Shoenfeld  6 , O Shovman  1   4   7
Affiliations

Evaluation of a machine learning tool for the early identification of patients with undiagnosed psoriatic arthritis - A retrospective population-based study

J Shapiro et al. J Transl Autoimmun. .

Abstract

Background: Psoriatic arthritis (PsA), an immune-mediated chronic inflammatory skin and joint disease, affects approximately 0.27% of the adult population, and 20% of patients with psoriasis. Up to 10% of psoriasis patients are estimated for having undiagnosed PsA. Early diagnosis and treatment can prevent irreversible joint damage, disability and deformity. Questionnaires for screening to identify undiagnosed PsA patients require patient and physician involvement.

Objective: To evaluate a proprietary machine learning tool (PredictAI™) developed for identification of undiagnosed PsA patients 1-4 years prior to the first time that they were suspected of having PsA (reference event).

Methods: This retrospective study analyzed data of the adult population from Maccabi Healthcare Service between 2008 and 2020. We created 2 cohorts: The general adult population ("GP Cohort") including patients with and without psoriasis and the Psoriasis cohort ("PsO Cohort") including psoriasis patients only. Each cohort was divided into two non-overlapping train and test sets. The PredictAI™ model was trained and evaluated with 3 years of data predating the reference event by at least one year. Receiver operating characteristic (ROC) analysis was used to investigate the performance of the model, built using gradient boosted trees, at different specificity levels.

Results: Overall, 2096 patients met the criteria for PsA. Undiagnosed PsA patients in the PsO cohort were identified with a specificity of 90% one and four years before the reference event, with a sensitivity of 51% and 38%, and a PPV of 36.1% and 29.6%, respectively. In the GP cohort and with a specificity of 99% and for the same time windows, the model achieved a sensitivity of 43% and 32% and a PPV of 10.6% and 8.1%, respectively.

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

Keywords: Artificial intelligence; Early diagnosis; Machine learning; Psoriasis; Psoriatic arthritis.

PubMed Disclaimer

Conflict of interest statement

Shapiro J, Cohen SB, Ben-Tov A, Shoenfeld Y and Shovman O are consultants for Predicta Med Analytics Ltd. All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.Getz B and Steinberg-Koch S have a published patent on A METHOD OF EVALUATING AUTOIMMUNE DISEASE RISK AND TREATMENT SELECTION, application number 17/596,015, however, the paper is referring to a specific use case of PsA disease.

Figures

Fig. 1
Fig. 1
Time gap model for training PredictAI™.
Fig. 2
Fig. 2
STARD flow diagram.
Fig. 3
Fig. 3
PPV vs Sensitivity, specificity cutoff selection 1 year prior to the reference event.
Fig. 4
Fig. 4
A (Left)- ROC Plot for identification of undiagnosed PsA patients in the test GP cohort, 1–4 years prior the reference event. B (Right) ROC Plot for identification of undiagnosed PsA patients in the Test PsO cohort 1–4 years prior the reference event.
Fig. 5
Fig. 5
Prominent features contributing to identification of undiagnosed PsA patients within the GP cohort. The left side of the SHAP figure represents the control population and how each feature contributes to the model classifying a patient without PsA. The right side of the figure represents the case population and how each feature contributes to the model classifying a patient with PsA.
See this image and copyright information in PMC

References

    1. Alinaghi F., Calov M., Kristensen L.E., et al. Prevalence of psoriatic arthritis in patients with psoriasis: a systematic review and meta-analysis of observational and clinical studies. J. Am. Acad. Dermatol. 2019;80(1):251–265. doi: 10.1016/j.jaad.2018.06.027. e19. - DOI - PubMed
    1. Wilson F.C., Icen M., Crowson C.S., McEvoy M.T., Gabriel S.E., Kremers H.M. Incidence and clinical predictors of psoriatic arthritis in patients with psoriasis: a population-based study. Arthritis Rheum. 2009;61(2):233–239. doi: 10.1002/art.24172. - DOI - PMC - PubMed
    1. Love T.J., Zu Y., Zhang Y., et al. Obesity and the risk of psoriatic arthritis: a population-based study. Ann. Rheum. Dis. 2012;71(8):1273–1277. doi: 10.1136/annrheumdis-2012-201299. - DOI - PMC - PubMed
    1. Eder L., Hadad A., Rosen C.F., et al. The incidence and risk factors for psoriatic arthritis in patients with psoriasis: a prospective cohort study. Arthritis Rheum. 2016;68(4):915–923. doi: 10.1002/art.39494.42. - DOI - PubMed
    1. Prey S., Paul C., Bronsard V., et al. Assessment of risk of psoriatic arthritis in patients with plaque psoriasis: a systematic review of the literature. J. Eur. Acad. Dermatol. Venereol. 2010;24(Suppl 2):31–35. doi: 10.1111/j.1468-3083.2009.03565. - DOI - PubMed