AI Tool Validates Eczema Severity Scoring with Smartphone Images

Overview

• A novel AI model developed by Japanese researchers uses smartphone images to objectively assess eczema severity, addressing gaps in self-reported symptoms.
• Validation studies published in *Allergy* show the AI-based TIS scoring system correlates strongly with dermatologist-assessed TIS (R=0.73) and meaningfully with SCORAD (R=0.53).
• The model integrates body part detection, lesion detection, and severity scoring algorithms trained on over 57,000 patient-shared images from Atopiyo LLC's dataset.
• Researchers identified a weak correlation between AI-derived severity scores and self-reported itch scores, highlighting discrepancies between perceived symptoms and clinical findings.
• Future developments aim to enhance the model's applicability by incorporating diverse skin types, age groups, and additional scoring systems like EASI and SCORAD.