A Brief Survey on Image Denoising based Feature Extraction and Classification Models for Oral Cancer Detection

A subset of oral cancer is called Oropharyngeal Carcinoma (OPC). As opposed to other forms of oral cancer, this one seems to be fatal more often. Those with oropharyngeal cancer also have damage to their tonsils, the back of their tongues, and the vallecula. The oral cavity is frequently attacked by malignant tumours, and oral cancer is one of the most common types. Understanding the precise nature of the oral cancer at an early stage allows for more precise treatment planning and more accurate prediction of the patient’s response to therapy. Diagnosing oral cancer from histological pictures is a significant obstacle. Visual characteristics of oral cancers are typically gleaned from epithelial layer architecture and the presence of keratin pearls. Salivary glands, tonsils, and the jawline, face, and mouth are all affected by the tumour. In this research, a brief survey is presented on image denoising models, feature extraction models for accurate oral cancer detection. A deep learning algorithm and cutting-edge technology make it possible to discover and classify things at an early stage. Multiple filtering strategies are used here to get rid of the speckle noise in the CT image. Oropharyngeal carcinoma is often detected with CT scans. In this research, a deep learning systems are analyzed for application in the classification of oral pictures into normal and pathological categories. By employing morphological operations, users can separate out the cancerous areas for accurate detection of oral cancer.