Convolutional neural network for oral cancer detection combined with improved tunicate swarm algorithm to detect oral cancer

Abstract Early Diagnosis of oral cancer is very important and can save you from some oral malignancies. However, while this approach aids in the rapid healing of patients and the preservation of their lives, there are several causes for poor and wrong diagnosis of oral cancer. In recent years, the use of computer-aided design diagnosis tools as an auxiliary tool alongside clinicians has greatly benefited in more accurate identification of this malignancy. The current study proposes a new approach for identifying oral cancer patients based on image processing and deep learning. The current study employs a recently integrated model of an improved tunicate swarm algorithm to produce an efficient tool for improving a convolutional neural network and delivering an accurate cancer diagnostic system. The approach is then implemented on the oral cancer pictures dataset. The approach is then validated by comparing it to other published papers using various measurement markers. The proposed model achieved an accuracy of 98.70% and a recall of 93.71% in detecting oral cancerous lesions from photographic images. The model also achieved an F1-score of 90.08% and a precision of 96.42%. The final results demonstrate that the offered approach can produce more exact results and can be used in conjunction with clinicians to help in diagnosing oral cancer.