Artificial intelligence algorithms and three-dimensional volumetric rendering for basal cell carcinoma detection and tumour depth assessment in reflectance confocal microscopy–optical coherence tomography images: a pilot study

Abstract

Abstract The reflectance confocal microscopy (RCM)–optical coherence tomography (OCT) device has shown utility in detecting and assessing the depth of basal cell carcinoma (BCC) in vivo but is challenging for novices to interpret. Artificial intelligence (AI) applied to RCM-OCT could aid readers. We trained AI models, using OCT rasters of biopsy-confirmed BCC, to detect BCC, create three-dimensional rendering and automatically measure tumour depth. Trained AI models were applied to a separate test set containing rasters of BCC, benign lesions, and healthy skin. Blinded reader analysis and tumour depth correlation with histopathology were conducted. BCC detection improved from viewing OCT rasters only (sensitivity 73.3%, specificity 45.5%) to viewing rasters with AI-generated BCC rendering (sensitivity 86.7%, specificity 48.5%). A Pearson correlation r2 = 0.59 (P = 0.02) was achieved for the tumour depth measurement between AI and histological measured depths. Thus, addition of AI to the RCM-OCT device may expand its utility widely.