Oral Vision: Advanced Image Recognition for Oral Cancer

Cancer that affects the lining of the mouth, lips, or upper throat is called oral cancer. Other names for it include tongue cancer, mouth cancer, and oral cavity cancer. There is an urgent need for early cancer detection due to the rise in cancer cases in all age groups. Alcohol consumption and tobacco usage are risk factors. The risk of oral cancer is fifteen times higher for people who use both alcohol and tobacco than for people who do not. This paper presents a deep learning-based technique that uses the ResNet50 convolution neural network model to classify the images of malignant and non-malignant oral cell images. To enhance the pretrained ResNet50 model's capacity to differentiate malignant and benign samples, it was trained using a dataset of histopathological oral tissue pictures. A special virtual voice assistant has also been integrated along with the model to make it more assistive for all kinds of users. In order to track the advancement of these methods, this paper also analyses how combining NLP with point-of-care tools makes use of a variety of algorithms and sets.

I. Introduction

Oral cancer is a significant global health issue affecting the tissues of the mouth, lips, and upper throat, often termed as oral cavity cancer or, depending on the area, tongue cancer. This form of cancer is particularly prevalent in regions with high tobacco and alcohol usage, and recent studies show that the simultaneous use of these substances can elevate the risk of oral cancer up to fifteen-fold. Early detection is essential in combating the increasing incidence of oral cancer across all age groups, as early-stage diagnosis is strongly linked to improved survival rates and quality of life post-treatment.