AI & Atopic Dermatitis: Severity Classification
- A new study suggests artificial intelligence could play a valuable role in objectively measuring the severity of atopic dermatitis (AD), also known as eczema. The research, published in...
- Researchers noted that smartphones and social media have made it easier for patients to document their condition.
- Researchers created three algorithms to analyze user-generated photos.
AI is revolutionizing how we assess atopic dermatitis. The research highlighted in this article showcases how an AI algorithm can now measure eczema severity objectively using images.This advancement could dramatically improve treatment, offering a more concrete metric than self-reported itch intensity, even though the study indicates patient perceptions and AI scores don’t always perfectly align. The AI model accurately detects body parts and eczema areas, paving the way for a new era in eczema management. Discover how News Directory 3 is covering the latest in medical breakthroughs and what this means for those affected by this chronic condition. Find out what the future holds in the ongoing journey to refine AI’s role in healthcare.
AI Model measures Eczema Severity, But Itch Perception Differs
Updated May 30, 2025
A new study suggests artificial intelligence could play a valuable role in objectively measuring the severity of atopic dermatitis (AD), also known as eczema. The research, published in the journal Allergy, found that an AI algorithm can successfully score eczema severity based on photos. However, the AI scores did not always align with how patients reported their itch intensity.
The chronic nature of AD requires ongoing management. Researchers noted that smartphones and social media have made it easier for patients to document their condition. In japan, over 28,000 users have shared more than 57,000 photos and comments about their symptoms on the Atopiyo platform.
Researchers created three algorithms to analyze user-generated photos. One algorithm detected body parts, another identified lesions, and the third assessed lesion severity using the Three Item Severity (TIS) scoring system, which rates severity from 0 to 9.
To validate the AI findings, investigators used the SCORing atopic Dermatitis (SCORAD) assessment, which includes both objective and subjective measures. Users of the AI algorithm also rated their itch intensity using the Itch-NRS-5 scoring system when uploading photos.
Analyzing 9,656 images from 900 participants, the AI model showed 98% accuracy in detecting body parts and 100% accuracy in detecting eczema areas.
In 220 images where patients also had their lesions assessed by a doctor,the AI-generated TIS scores correlated well with physician scores. However, when analyzing 8,556 images, the model’s severity scores showed a weak correlation with patient-reported itch scores. This suggests that disease severity and itch do not always correspond.
The authors explained that the divergence between subjective and objective measures highlights the need for more precise assessment methods, suggesting AI could offer a solution.
“The AI model developed in this study has the potential to help patients with AD objectively assess their skin condition, facilitating timely and appropriate treatment,” they wrote.
The researchers added that their models need validation using a more diverse data set,including people with different skin types.
What’s next
Further research will focus on validating the AI model across diverse populations to ensure its broad applicability in objectively measuring eczema severity and guiding treatment decisions.