AI Revolutionizing Psoriasis Severity Assessment for Precise Therapy and QoL Improvement

Psoriasis (PsO) is a chronic and incurable immune-mediated skin disease affecting about 125 million individuals globally. It imposes a significant economic burden due to treatment costs and related functional impairments, and traditional diagnosis methods can lead to comorbidities and reduced quality of life (QoL).

In a study published in the Journal of Medical Internet Research, researchers emphasize that precise therapy and long-term management are crucial to reduce comorbidities and enhance patients' QoL. PsO severity guides treatment decisions, necessitating an accurate and objective assessment tool. The Psoriasis Area and Severity Index (PASI) is widely used but suffers from subjectivity and poor consistency among clinicians. To help address these issues, the study aimed to develop a more objective and intelligent approach to assess psoriasis severity using artificial intelligence (AI) to see if this approach has any potential benefits.

With approval from the Clinical Research Ethics Committee, the researchers gathered data from psoriasis patients between 2015 and 2021 and created a dataset of 14,096 images from 2,367 patients. Researchers divided the data into training and validation sets. Experienced clinicians assessed the images, and the PASI scores were averaged. The AI model incorporated a multiview deep feature enhancement block to mimic clinical PASI assessment, aiming for improved objectivity and consistency, and was integrated into the SkinTeller mobile app designed to assess psoriasis severity accurately and recommend personalized treatment.

The AI-based system outperformed 43 experienced dermatologists in assessing psoriasis severity. The SkinTeller app, integrated with this system, was used 3,369 times by patients from 18 hospitals, with feedback from 43 dermatologists confirming its excellent performance. Confirming a promising alternative for precise assessment and self-management of PsO.

"The SkinTeller app integrated with the proposed system show enormous potentialities for precise personalized treatment and chronic disease self-management in patients with psoriasis," said researchers.

Reference

Huang K, Wu X, Li Y, et al. Artificial intelligence–based psoriasis severity assessment: real-world study and application. Journal of Medical Internet Research. 2023;25:e44932. doi:10.2196/44932