Detection and Classification of Oral Cancer Using Machine Learning Models

Around 177,757 people worldwide lose their lives to oral cancer each year, the most prevalent type of head and neck cancer. A 75–90% chance of survival for oral cancer is achieved with early identification. However, the majority of cases are discovered in advanced stages because of a lack of public awareness of mouth cancer symptoms and delays in referrals to specialists. The prognosis of oral cancer can still be improved with early detection and treatment. Oral cancer screening now has more options thanks to the development of vision-based supplemental technologies that can detect oral potentially malignant disorders (OPMDs), which are linked to the advancement of cancer. In this study, a machine learning-based model for identifying and classifying oral cancer is developed and tested. The Kaggle dataset utilized in this work is a binary classification task that separates oral tissue samples into carcinogenic (class 1) and non-cancerous (class 0) conditions. The research explores the intricate connection between ambiguity and burstiness in text, drawing comparisons to the challenging detection challenges encountered by oral cancer.