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.
ABSTRACT
Background: The therapeutic interventions for basal cell carcinoma (BCC) impact facial cosmesis. Objectives: Our aim was to assess the aesthetic burden of facial BCC treatment by evaluating the extent of the tumour site visually based on clinical images before and after immunocryosurgery; a minimally invasive combination treatment of topical imiquimod and cryosurgery. Materials & Methods: A three-item (texture, height, colour) burn scar scale was independently applied by four physicians (two dermatologists and two plastic surgeons) on archival semi-standardized clinical images of facial BCC before and one year after immunocryosurgery. The score assessments were compared using non-parametric statistical tests; internal consistency (reliability) and inter-rater agreement were assessed using Cronbach's α and Gwet's AC2, respectively. Results: Images (before and one year after treatment) of 27 BCCs from 26 patients (15 males) were analysed. The reliability was good (α>0.80) for all items before surgery and for colour after surgery. The inter-rater agreement was acceptable (AC2>0.70) for all items except for height pre-treatment (AC2 = 0.482). Based on averaging of all raters, cosmesis improved significantly after treatment at all tumour sites (p<0.001, Wilcoxon signed ranks test). The main limitations of the study are the inclusion of cases from a single centre and its retrospective nature. Conclusion: The evaluated burn scar scale can be reliably used to compare the extent of the BCC site visually before therapy and after resultant scars have formed. In this way, the impact of therapeutic interventions on cosmesis (from tumour to scar) can be evaluated, particularly for less aggressive facial skin tumours, such as most BCC.
