Oral Cancer Stage Classification Using Machine Learning

Abstract

Oral cancer (OC)is frequently occurring cancer that affects various areas within the oral cavity. Despite the advancement of sophisticated diagnostic and therapeutic techniques, the morbidity and mortality rates associated with oral cancer continue to be concerning. OC can affect a person’s quality of life. Beforehand identification of oral cancer can greatly reduce the need for aggressive treatment. In recent times machine learning ways has shown promising results in the identification of oral cancer. In this study, we have applied machine learning techniques such as SVM, XG Boost, and Random Forest (RF) in the stage classification of oral cancer. Our dataset contains a collection of oral images obtained from different levels of oral cancer patients from SMS hospital. We extracted various features from these images, such as color, texture, and shape, and used them as input data in our machine-learning model. We have calculated various measures like accuracy, precision, and F1 score and found that accuracy Random Forest have outperformed than SVM and XG Boost by achieving accuracy as 87.1%.