Correlation of Dermoscopic and Histopathological Features in Basal Cell Carcinoma Using Computerized Image Analysis

Background: Basal cell carcinoma (BCC) is the most common skin cancer, exhibiting local invasiveness despite its low metastatic potential. Dermoscopy and histopathology are essential for diagnosis, while quantitative assessments may enhance lesion characterization.Study: This study aims to analyze the dermoscopic and histopathological characteristics of BCC and investigate the correlation between dermoscopic pigmentation patterns and tumor depth to improve lesion classification and diagnostic accuracy. Patients and Methods: This retrospective study analyzed 41 patients with 42 histopathologically confirmed BCC lesions, evaluated at Nizip State Hospital and 25 Aralık State Hospital between April 2023 and February 2025. High-resolution dermoscopic images were analyzed alongside histopathological findings. AI-assisted computerized image analysis was employed to quantify lesion size and pigmentation percentage, while tumor depth and dermoscopic-histopathological correlations were manually assessed.Results: BCC was more prevalent in males (56.1%) and older adults, with a mean age of 67.1 years. The most commonly affected site was the nose (42.9%), followed by the cheek (14.3%) and upper lip (11.9%). Histopathologically, nodular (28.6%) and adenoid (28.6%) BCC were the most frequent subtypes. Dermoscopic analysis revealed blue-gray ovoid nests (57.14%) and arborizing telangiectasias (71.43%) as predominant features, particularly in mixed-type BCC, while blue-gray dots and globules (57.14%) were most common in micronodular BCC. Ulceration (45.24%) and multiple erosions (57.14%) were strongly associated with infiltrative BCC. A negative correlation was observed between pigmentation percentage and tumor depth, with deeper tumors exhibiting reduced pigmentation, though this trend was not statistically significant.Comprehensive characterization of the dermoscopic and histopathological features of BCC enhances lesion differentiation. AI-assisted lesion size and pigmentation analysis, combined with histopathological evaluation, improves diagnostic precision. Further studies with larger cohorts are needed to validate these findings and refine classification criteria.