Analysis of Oral Cancer Detection based Segmentation and Classification using Deep Learning Algorithms
Pullaiah Pinnika, K. Venkata Rao · Advances in computer science research · 2024
Oral cancer is deadly cancer which is majorly spread in less and middle-income countries.The early diagnosis of oral cancer may attained through automatic detection of cancerous and malignant mouth lesions.Various researches developed a Machine Learning (ML) method which detects oral cancer from images.Though, there still lack in huge precision in the detection of oral cancer.Recently, development of Artificial Intelligence (AI), Deep Learning (DL) algorithms effectively detects the oral cancer in early and maximizes a patient's survival rate.This survey analysis different DL algorithms such as Modified K-Means and Fuzzy C-means (modified KFCM), UNet depended Bayesian Deep Learning (BDL), Capsule network and CariesNet which was used for segmentation of oral cancer.Then, AlexNet, Enhanced Grasshopper Optimization Algorithm (EGOA) depended Deep Belief Network (DBN), Convolutional Neural Network (CNN) and Deep CNN was used for classification of oral cancer.The performance metrics used for evaluating the algorithms are Dice coefficient, Jaccard, Accuracy, Mean IoU, Precision, Weighted IoU, Specificity, Recall, Sensitivity, Error rate, F1-score and AUC.