Computer-Guided PASI Score Technology Comparable to Physician Assessment

pustular psoriasis
pustular psoriasis
Researchers performed automated total body imaging and digital image analysis to create an objective, computer-guided technology to evaluate patients' PASI score.

An automated, computer-guided Psoriasis Area and Severity Index (PASI) measurement technology shows comparable precision and excellent reproducibility at analyzing psoriatic arthritis severity when compared with a physicians’ assessment, according to data published in The British Journal of Dermatology.

Researchers performed automated total body imaging and digital image analysis to create an objective, computer-guided technology to evaluate patients’ PASI score. Patients with plaque-like psoriasis (N=120) were evaluated by both the automated, computer-guided technology and by 3 PASI-trained physicians for levels of agreement, variability between methods, and reproducibility of each assessment.

Overall, the mean PASI score was 8.8, but there was a wide range in severity from 0.7 to 43.8. The 3 physicians calculated a slightly lower PASI score when compared with the automated computer-guided technology, which lead to an intraclass correlation coefficient of 0.86 (95% CI, 0.8-0.9; P <.001). The physicians’ reproducibility was high, with an inter-observer intraclass correlation coefficient of 0.9 (95% CI, 0.86-0.93; P <.001). The automated, computer-guided technology’s reproducibility was very high with an intraclass correlation coefficient of 0.99 (95% CI, 0.98-0.99; P <.001).

Due to the computer analyzing process, exact body postures are needed for imaging, which could be limiting for patients who are immobile or non-cooperative.

Overall, this method could provide objective data useful in clinical trials or conclusive documentation of disease severity.

Related Articles

Follow @DermAdvisor

Reference

Fink C, Alt C, Uhlmann L, Klose C, Enk A, Haenssle HA. Precision and reproducibility of automated computer-guided Psoriasis Area and Severity Index measurements in comparison with trained physicians [published online September 15, 2018]. Br J Dermatol. doi: 10.1111/bjd.17200