Comparative Assessment of Dermoscopic Patterns in Early Melanoma versus Benign Pigmented Lesions: A Diagnostic Accuracy Study

Background: Early detection of melanoma remains critical for improving patient survival outcomes. Dermoscopy has emerged as an essential non-invasive diagnostic tool for differentiating malignant from benign pigmented lesions. However, distinguishing early melanoma from benign melanocytic nevi presents significant diagnostic challenges. This study aimed to evaluate the diagnostic accuracy of specific dermoscopic patterns in differentiating early melanoma from benign pigmented lesions. Methods: This prospective diagnostic accuracy study was conducted at a dermatology referral center. A total of 524 pigmented lesions from 489 patients were evaluated using standardized dermoscopic examination. All lesions underwent histopathological examination as the reference standard. Dermoscopic features were assessed using established criteria, and diagnostic performance metrics were calculated. Results: The study included 147 early melanomas (Breslow thickness ≤1.0 mm) and 377 benign pigmented lesions. Atypical pigment network demonstrated the highest sensitivity (87.8%) for melanoma detection, while blue-white veil showed the highest specificity (94.7%). The combination of three or more melanoma-specific structures yielded sensitivity of 91.2%, specificity of 89.4%, and area under the curve (AUC) of 0.934 (95% CI: 0.912-0.956). Irregular streaks (OR=8.42, p<0.001), atypical dots/globules (OR=6.78, p<0.001), and regression structures (OR=5.94, p<0.001) were independently associated with melanoma diagnosis. The seven-point checklist achieved sensitivity of 85.7% and specificity of 82.5%. Conclusion: Systematic dermoscopic assessment using multiple melanoma-specific features significantly improves diagnostic accuracy for early melanoma detection. Integration of pattern analysis with algorithmic approaches optimizes differentiation between early melanoma and benign pigmented lesions.