AI Eczema Severity: Smartphone App
- A new artificial intelligence (AI) tool can assess the severity of atopic dermatitis (AD), or eczema, using photos taken with a smartphone.
- The AI uses three integrated algorithms to detect body parts,identify skin lesions,and assess severity based on the Three-Item Severity (TIS) score.
- The AI model was trained using a large database from Atopiyo, an online AD platform where over 28,000 users share more than 57,000 photos and comments about their...
An innovative AI tool accurately assesses eczema severity using smartphone photos, offering a breakthrough for atopic dermatitis (AD) sufferers. News Directory 3 reports on this new model, which utilizes advanced algorithms to detect body parts and identify areas affected by the skin condition with remarkable precision. Japanese researchers developed this AI to correlate with clinical assessments, potentially revolutionizing AD management by providing an objective measure of severity. With the AI model showing impressive accuracy in detecting affected areas, the potential for improved patient monitoring and clinical assessments is immense. This digital assessment could significantly complement traditional methods. Discover what’s next as researchers work to include a broader scope within the AI.
AI Rates Eczema Severity From Smartphone Snaps
Updated June 9, 2025
A new artificial intelligence (AI) tool can assess the severity of atopic dermatitis (AD), or eczema, using photos taken with a smartphone. Developed by Japanese researchers, the AI model offers an objective way to evaluate the chronic skin condition.
The AI uses three integrated algorithms to detect body parts,identify skin lesions,and assess severity based on the Three-Item Severity (TIS) score. The TIS score evaluates erythema, edema or papulation, and excoriation on a scale of 0-9.
The AI model was trained using a large database from Atopiyo, an online AD platform where over 28,000 users share more than 57,000 photos and comments about their symptoms. Researchers trained the AI using 880 images and tested it with 220 images. The model accurately detected 98% of body parts and all eczema-affected areas.
The AI’s outputs were compared with established clinical scoring systems, including scoring Atopic Dermatitis (SCORAD). A study published in the journal Allergy showed a strong correlation between the AI-TIS and objective measures, supporting its potential for clinical use.
While the AI-TIS showed a strong correlation with objective measures,it had a weaker correlation with patient-reported itch severity. This suggests that itch severity does not always align with the visible severity of eczema.
The AI tool’s ability to quantify objective signs of atopic dermatitis from home photographs could improve the precision of clinical assessments and enhance daily patient monitoring.
What’s next
Researchers plan to expand the AI model to include a wider age range, diverse skin types, and elements from other clinical scales. This advancement could lead to better patient care and clinical research in dermatology.