Researchers have created a new artificial intelligence (AI) model to assess the severity of eczema using photos uploaded by patients. The model showed strong accuracy in diagnosing the condition.
Takeya Adachi from Keio University School of Medicine and his team worked with Atopiyo, LLC to develop this tool. Their goal was to provide a more standardized and objective way to evaluate eczema severity.
They explained in the journal *Allergy* that while social media and smartphone apps help patients track symptoms and learn about their condition, differences often exist between patient reports and clinical assessments. This gap highlights the need for reliable evaluation methods, where digital tools could play an important role.
The team accessed over 57,000 images shared by more than 28,000 users since 2018. They trained three AI algorithms: one to identify body parts, another to detect skin lesions, and a third to assess severity based on the Three Item Severity (TIS) scale, which measures redness, swelling, and skin damage.
During testing, the AI model analyzed 880 images and achieved high diagnostic accuracy. In a follow-up study involving 220 patients, the AI’s severity scores strongly matched those given by dermatologists (correlation coefficient R = 0.73, statistically significant). The AI’s scores also showed a meaningful link with objective SCORAD scores, another clinical measure of eczema severity (R = 0.53).
Dr. Adachi said in a press release, “Many patients with eczema struggle to evaluate their disease severity on their own. Our AI model allows for objective, real-time tracking using just a smartphone, empowering patients and potentially improving disease management.”
This advancement could help patients monitor their eczema more accurately and support better treatment decisions.
Related Topics:
