iDScore

iDScore enhances diagnostic classification of melanocytic skin lesions by integrating patient clinical data and dermoscopic features to distinguish early melanoma (EM) from atypical nevi (AN).

Key Features:

• Integrated Data Approach: Combines patient-specific clinical variables (age, sex, lesion diameter, body site) with dermoscopic imaging features.
• Multicenter Image Dataset: Validated on a multicenter dataset from European dermatology centers comprising 904 dermoscopic images.
• Diagnostic Accuracy: In trials of 904 images evaluated by 111 participants, iDScore achieved an average Area Under the Curve (AUC) of 0.776 across participant sessions.
• Skill Level Adaptability: Improved diagnostic accuracy by 10–16% across users compared with intuitive diagnosis, the ABCD rule, and the seven-point checklist.
• Concordance and Decision Support: Increased inter-user concordance and reduced incorrect management decisions in evaluations.

Scientific Applications:

• Early Melanoma Detection: Supports early detection of melanoma (EM) by improving differentiation from atypical nevi (AN) in dermoscopic assessments.
• Differential Diagnosis of Atypical Nevi: Assists distinction of atypical nevi from early melanoma and other ambiguous melanocytic skin lesions (MSLs).
• Multicenter Study Standardization: Provides a standardized scoring method for multicentric studies evaluating MSLs across different clinical environments.

Methodology:

iDScore uses a scoring classifier that integrates patient clinical variables (age, sex, lesion diameter, body site) with dermoscopic features.