Image Perceptual Similarity Metrics for the Assessment of Basal Cell Carcinoma

SIMPLE SUMMARY: The impact of basal cell carcinomas (BCCs) on a patient’s appearance can be significant. Reliable assessments are crucial for the effective management and evaluation of therapeutic interventions. Given that color and texture are critical attributes that describe the clinical aspect of skin lesions, our focus was to devise metrics that capture the way experts perceive deviations of target BCC areas from the surrounding healthy skin. Using computerized image analysis, we explored various similarity metrics to predict perceptual similarity, including different color spaces and distances between features from image embeddings derived from a pre-trained deep convolutional neural network. The results are promising in providing a valid, reliable, and affordable modality, enabling more accurate and standardized assessments of BCC tumors and post-treatment scars. Our approach to modeling color and texture lesion similarity from the surrounding healthy skin is a promising paradigm for the further development of a valid and reliable scar assessment tool. ABSTRACT: 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.