Efficient management of basal cell carcinomas (BCC) requires reliable assessments of both tumors and post-treatment scars. We aimed to estimate image similarity metrics that account for BCC’s perceptual color and texture deviation from perilesional skin. In total, 176 clinical photographs of BCC were assessed by six physicians using a visual deviation scale. Internal consistency and inter-rater agreement were estimated using Cronbach’s α, weighted Gwet’s AC2, and quadratic Cohen’s kappa. The mean visual scores were used to validate a range of similarity metrics employing different color spaces, distances, and image embeddings from a pre-trained VGG16 neural network. The calculated similarities were transformed into discrete values using ordinal logistic regression models. The Bray–Curtis distance in the YIQ color model and rectified embeddings from the ‘fc6’ layer minimized the mean squared error and demonstrated strong performance in representing perceptual similarities. Box plot analysis and the Wilcoxon rank-sum test were used to visualize and compare the levels of agreement, conducted on a random validation round between the two groups: ‘Human–System’ and ‘Human–Human.’ The proposed metrics were comparable in terms of internal consistency and agreement with human raters. The findings suggest that the proposed metrics offer a robust and cost-effective approach to monitoring BCC treatment outcomes in clinical settings.
Background. Complete basal cell carcinoma (BCC) excision remains the most common treatment modality. However, its clinical characteristics and the surgical outcomes achieved in patients over 80 years—often with several medical comorbidities and potentially limited life expectancy—have not been thoroughly examined. This clinical study aims to investigate tumor-specific characteristics and surgical outcomes following surgical treatment of BCC in older individuals. Methods. An observational cohort study based on a prospectively maintained database was conducted in a tertiary center using a predetermined protocol. Patients who underwent BCC surgery between January 2010 and September 2024 were included and grouped by age under or over 80 years. The inclusion criterion was a histologically confirmed BCC, while patients with syndromes predisposing BCC development were excluded. Results. Among the 1396 biopsy-proven BCCs, 35% of the patients were older than 80 years. No significant differences were observed in their baseline characteristics. The pathogenic capacity was greater in elderly patients, who exhibited higher rates of multiple and concurrent skin cancers, larger BCC diameters, and routine involvement in high-risk areas. More lesions were classified as high-risk for recurrence, and the surgical treatment was accompanied by a higher frequency of positive or close margins, high-grade subtypes, and perineural invasion. Logistic regression of 1150 BCCs revealed that age > 80, advanced TNM stage, and margin status robustly predict high-risk histology and high NCCN risk of tumor recurrence. Conclusions. This study highlights that BCC in the elderly population tends to present with a more aggressive tumor status, based on the key clinical and pathology features. These findings underscore the need for tailored surgical strategies in this population.
Efficient management of basal cell carcinoma (BCC) requires reliable assessments of both tumors and post-treatment scars. We aimed to estimate image similarity metrics that account for BCC's perceptual color and texture deviation from perilesional skin. 176 clinical photographs of BCC were assessed by six physicians using a visual deviation scale. Internal consistency and inter-rater agreement were estimated using Cronbach's α, weighted Gwet's AC2, and quadratic Cohen's Kappa. The mean visual scores were used to validate a range of similarity metrics employing different color spaces, distances, and image embeddings from pre-trained VGG16 neural network. The calculated similarities were transformed into discrete values using ordinal logistic regression models. The Bray-Curtis distance in the YIQ color model and rectified embeddings from the 'fc6' layer minimized the mean squared error and demonstrated strong performance in representing perceptual similarities. Box plot analysis and Wilcoxon rank-sum test were used to visualize and compare the levels of agreement, conducted on a random validation round between the two groups: ‘Human-System’ and ‘Human-Human’. The proposed metrics were comparable in terms of internal consistency and agreement with human raters. The findings suggest that the proposed metrics offer a robust, and cost-effective approach to monitoring BCC treatment outcomes in clinical settings.