Generative Artificial Intelligence Tools: Evaluating Ways to Automate Your SALT (GATEWAYS) Scoring of Alopecia Areata

March 2026 in “ Pediatric Dermatology ”

Radhika Gupta, Hojjat Salmasian, Michelle Oboite, Sarah Mehlmann, Griffin Stockton Hogrogian, Leslie Ann Castelo‐Soccio, George Cotsarelis, Albert C. Yan

alopecia areata Severity of Alopecia Tool SALT generative AI GPT-4o image assessments intraclass correlation coefficients concordance correlation coefficients AI Children's Hospital of Philadelphia

TLDR Generative AI tools can accurately score alopecia areata, reducing subjectivity in evaluations.

This study evaluated the use of GPT-4o, a generative AI tool, to automate the scoring of alopecia areata using the Severity of Alopecia Tool (SALT) based on image assessments. Conducted at the Children's Hospital of Philadelphia, the study involved 104 de-identified 4-view image sets of patients' scalps. The AI-generated SALT scores were compared to those provided by in-person and image-based assessments by providers. The results showed high concordance, with intraclass correlation coefficients (ICC) and concordance correlation coefficients (CCC) ranging from 0.815 to 0.866 for AI versus in-person assessments, and 0.833 to 0.817 for AI versus image-based assessments. These findings suggest that generative AI tools like GPT-4o could serve as valuable adjuncts in the objective scoring of alopecia areata, potentially reducing subjectivity in clinical evaluations.

View this study on onlinelibrary.wiley.com →

Discuss this study in the Community →

Research cited in this study

6 / 6 results

research Using Artificial Intelligence to Compute Severity of Alopecia Tool Scores

5 citations , April 2024 in “JAAD International”

AI can accurately measure hair loss severity in alopecia areata.

research Two-Dimensional Planimetry for Alopecia Areata Severity Evaluation Compared With Severity of Alopecia Tool: A Pilot Study

August 2023 in “Skin Research and Technology”

Measuring bald patch size can help grade hair loss severity, with photograph-based evaluation being more reliable.

‘You Lose Your Hair, What’s the Big Deal?’ I Was So Embarrassed, I Was So Self-Conscious, I Was So Depressed: A Qualitative Interview Study to Understand the Psychosocial Burden of Alopecia Areata

research ‘You Lose Your Hair, What’s the Big Deal?’ I Was So Embarrassed, I Was So Self-Conscious, I Was So Depressed: A Qualitative Interview Study to Understand the Psychosocial Burden of Alopecia Areata

30 citations , September 2020 in “Journal of Patient-Reported Outcomes”

Alopecia Areata (AA) causes significant emotional distress, including feelings of embarrassment, depression, and anxiety, and impacts social interactions and daily activities.

Automated Digital Image Analysis (TrichoScan) in Male Patients with Androgenetic Alopecia: Comparison with Manual Marking of Hairs on Trichoscopic Images

research Automated Digital Image Analysis (TrichoScan) in Male Patients with Androgenetic Alopecia: Comparison with Manual Marking of Hairs on Trichoscopic Images

5 citations , January 2018 in “Skin Research and Technology”

TrichoScan needs optimization as it underestimated hair density by 38.9% compared to manual counting.

research Quantifying Alopecia Areata via Texture Analysis to Automate the SALT Score Computation

20 citations , December 2017 in “Journal of Investigative Dermatology Symposium Proceedings”

Researchers created a fast, accurate computer program to measure hair loss in alopecia areata patients.

research Objective Outcome Measures: Collecting Meaningful Data on Alopecia Areata

39 citations , November 2017 in “Journal of The American Academy of Dermatology”

The document suggests using standardized methods to track and measure hair loss in alopecia areata, including patient self-assessment and a 50% improvement in specific scores as a treatment goal.