A Remote Network Transmission Diagnosis Method for Oral Cancer Based on 6G and Rough Set Theory Hierarchical Diagnosis

This paper presents a novel Deep Fuzzy Convolutional Neural Network remote oral cancer diagnosis (DFCNN-OC) technique to improve remote diagnostic procedures for oral cancer by using the unique capabilities of 6G telecommunications and Rough Set Theory (RST) for hierarchical diagnosis. The proposed DFCNN framework combines fuzzy logic principles into the convolutional neural network architecture to address the problematic structures and variations in oral cancer images. This combination improves the stability and accuracy of the model by making it easier to handle doubts and errors in medical pictures. The DFCNN model uses the outstanding speed and reliability of 6G technology to guarantee the efficient and fast transfer of large amounts of oral cancer image data in real-time and allow quick and exact remote diagnosis. The proposed methodology includes highlighting areas of interest in the images through preprocessing and then improving the feature extraction process with fuzzy logic. This procedure can identify minute but essential details that traditional approaches fail to achieve. Then, using Rough Set Theory to make decisions, feature selection is improved to reduce difficulty without losing diagnostic accuracy. The classification step uses an advanced DFCNN to distinguish between benign and malignant diseases accurately. The DFCNN model outperformed other currently used methods when tested against the Oral Cancer (Lips and Tongue) Images (OCI) dataset. The efficacy of the DFCNN approach is effectively validated and highlights its ability to develop remote oral cancer diagnosis with low latency of 3.5 ms.