Convolutional Neural Network-Based Clinical Predictors of Oral Dysplasia: Class Activation Map Analysis of Deep Learning Results

Oral cancer/oral squamous cell carcinoma is among the top ten most common cancers globally, with over 500,000 new cases and 350,000 associated deaths every year worldwide. There is a critical need for objective, novel technologies that facilitate early, accurate diagnosis. For this purpose, we have developed a method to classify images as “suspicious” and “normal” by performing transfer learning on Inception-ResNet-V2 and generated automated heat maps to highlight the region of the images most likely to be involved in decision making. We have tested the developed method’s feasibility on two independent datasets of clinical photographic images of 30 and 24 patients from the UK and Brazil, respectively. Both 10-fold cross-validation and leave-one-patient-out validation methods were performed to test the system, achieving accuracies of 73.6% (±19%) and 90.9% (±12%), F1-scores of 97.9% and 87.2%, and precision values of 95.4% and 99.3% at recall values of 100.0% and 81.1% on these two respective cohorts. This study presents several novel findings and approaches, namely the development and validation of our methods on two datasets collected in different countries showing that using patches instead of the whole lesion image leads to better performance and analyzing which regions of the images are predictive of the classes using class activation map analysis.

Wnt/β-Catenin Signaling in Oral Carcinogenesis

Oral carcinogenesis is a complex and multifactorial process that involves cumulative genetic and molecular alterations, leading to uncontrolled cell proliferation, impaired DNA repair and defective cell death. At the early stages, the onset of potentially malignant lesions in the oral mucosa, or oral dysplasia, is associated with higher rates of malignant progression towards carcinoma in situ and invasive carcinoma. Efforts have been made to get insights about signaling pathways that are deregulated in oral dysplasia, as these could be translated into novel markers and might represent promising therapeutic targets. In this context, recent evidence underscored the relevance of the Wnt/β-catenin signaling pathway in oral dysplasia, as this pathway is progressively “switched on” through the different grades of dysplasia (mild, moderate and severe dysplasia), with the consequent nuclear translocation of β-catenin and expression of target genes associated with the maintenance of representative traits of oral dysplasia, namely cell proliferation and viability. Intriguingly, recent studies provide an unanticipated connection between active β-catenin signaling and deregulated endosome trafficking in oral dysplasia, highlighting the relevance of endocytic components in oral carcinogenesis. This review summarizes evidence about the role of the Wnt/β-catenin signaling pathway and the underlying mechanisms that account for its aberrant activation in oral carcinogenesis.