Oral cancer is a major global health issue accounting for 177,384 deaths in 2018 and it is most prevalent in low- and middle-income countries. Enabling automation in the identification of potentially malignant and malignant lesions in the oral cavity would potentially lead to low-cost and early diagnosis of the disease. Building a large library of well-annotated oral lesions is key. As part of the MeMoSA ® (Mobile Mouth Screening Anywhere) project, images are currently in the process of being gathered from clinical experts from across the world, who have been provided with an annotation tool to produce rich labels. A novel strategy to combine bounding box annotations from multiple clinicians is provided in this paper. Further to this, deep neural networks were used to build automated systems, in which complex patterns were derived for tackling this difficult task. Using the initial data gathered in this study, two deep learning based computer vision approaches were assessed for the automated detection and classification of oral lesions for the early detection of oral cancer, these were image classification with ResNet-101 and object detection with the Faster R-CNN. Image classification achieved an F1 score of 87.07% for identification of images that contained lesions and 78.30% for the identification of images that required referral. Object detection achieved an F1 score of 41.18% for the detection of lesions that required referral. Further performances are reported with respect to classifying according to the type of referral decision. Our initial results demonstrate deep learning has the potential to tackle this challenging task.

Oral Submucous fibrosis (OSMF) has traditionally been described as “a chronic, insidious, scarring disease of the oral cavity, often with involvement of the pharynx and the upper esophagus”. Millions of individuals are affected, especially in South and South East Asian countries. The main risk factor is areca nut chewing. Due to its high morbidity and high malignant transformation rate, constant efforts have been made to develop effective management. Despite this, there have been no significant improvements in prognosis for decades. This expert opinion paper updates the literature and provides a critique of diagnostic and therapeutic pitfalls common in developing countries and of deficiencies in management. An inter-professional model is proposed to avoid these pitfalls and to reduce these deficiencies.

Oral cancers (OC) are among the most frequent malignancies encountered in Southeast Asia, primarily due to the prevalent habit of betel quid (BQ) and smokeless tobacco use in this region. Areca nut (AN), the primary ingredient in BQ, contains several alkaloids, including arecoline, arecaidine, guvacoline, and guvacine. These have been associated with both the AN abuse liability and carcinogenicity. Additionally, variations in AN alkaloid levels could lead to differences in the addictiveness and carcinogenic potential across various AN-containing products. Recent studies based on animal models and in vitro experiments show cellular and molecular effects induced by AN. These comprise promoting epithelial-mesenchymal transition, autophagy initiation, tissue hypoxia, genotoxicity, cytotoxicity, and cell death. Further, clinical research endorses these undesired harmful effects in humans. Oral submucosal fibrosis, a potentially malignant disease of the oral cavity, is predominantly reported from the geographical areas of the globe where AN is habitually chewed. OC in chronic AN users presents a more aggressive phenotype, such as resistance to anti-cancer drugs. The available evidence on the carcinogenicity of AN based on the findings reported in the recently published experimental studies is discussed in the present review.

Areca nut is a highly addictive substance with carcinogenic properties and causes many harmful effects to the human body. Alkaloids are the major chemicals found in areca nut, and their mechanism of action has been studied for several years. About 600 million people around the world use areca nut, and its usage is higher in Asian countries. Areca nut chewers get used to the habit mainly due to social influence, stress, or lack of awareness. Some studies have reported a dependence syndrome related to areca nut chewing. Even though there are many ongoing research studies regarding tobacco addiction, less concern has been given to the areca nut addiction. There are only few studies related to areca nut cessation, and even those few cessation programs are mainly designed using tobacco cessation methodology. Sociocultural influence, behavioral factors, and accessibility are the main barriers to cessation. Behavior changing interventions proved to be more effective in areca nut cessation, but recently studies on pharmacological therapies are also emerging.

Oral Squamous Cell Carcinoma (OSCC) poses a significant health challenge, with early detection being crucial for effective treatment and improved survival rates. While previous studies have examined the use of standard photographs, such as those from smartphones, for oral lesion classification, they typically rely solely on images, overlooking the potential benefits of incorporating multiple modalities. This study addresses this gap by proposing a multimodal deep-learning pipeline incorporating diverse data sources, including patient metadata, which mimics the diagnostic approach of clinicians in early detection of oral cancer. The study leverages state-of-the-art image encoders to classify oral lesions into benign and potentially malignant categories. A performance comparison of six pre-trained deep-learning models (MobileNetV3-Large, MixNet-S, ResNet-50, HRNet-W18-C, DenseNet-121, and Inception_v3) is presented. The performance of the proposed pipeline achieved an overall accuracy of 81%, precision of 79%, recall of 79%, F1-score of 78%, and a Matthews Correlation Coefficient (MCC) of 0.57 using the MobileNetV3-Large image encoder. The findings highlight the efficacy of integrating multiple data modalities for more accurate early detection of potential malignancies compared to using only image data. The outcomes could pave the way for improved clinical decision-making and patient outcomes.