102 A deep learning system enables a prompt and objective scoring of Nail Psoriasis Severity Index

In the practice of psoriasis, nail lesions must be accurately evaluated in addition to skin and joint lesions. Scoring with Nail Psoriasis Severity Index (NAPSI) is often used as a tool assess severity but there are several vexing issues. First, calculating NAPSI scores on multiple digits rather time-consuming. Second, vary widely among scoring dermatologists. this study, we developed deep learning-based prediction system solve current problems make it easier more effective implement daily clinical practice. The “nail region detector” that detects nails images was constructed utilizing an object detection model. “NAPSI predictor” predicts detected using classification Those were combined integrated system. A total 1876 hand foot containing for collected from Google searches well our patients psoriasis. We prepared 2939 by manually cropping out regions patients. Annotation performed dermatologists abundant experience performance then comparison between results showed recognized 93.8% accuracy, proposed 82.8% accuracy. also able predict higher accuracy lower variability than Accordingly, shown learning can quickly isolate estimate at level comparable dermatologist. method study expected applicable not only psoriasis image analysis other diseases future investigations.