Millions of individuals are afflicted with lethal oral cancer. Early oral cancer identification improves treatment and survival. In recent years, convolutional neural networks (CNNs) have shown great potential for medical image processing applications, including cancer detection. This work focuses on using deep CNNs to identify oral cancer quickly. A large collection of oral cavity images is used in the recommended method. Integrating intraoral photos, radiographs, and histopathology slides to train and improve a deep learning CNN model. This model automatically detects and classifies oral cancerous tumors, including precancerous and malignant areas. Preprocessing photos, supplementing the dataset to improve model generalization, and using sophisticated neural architectures to extract and learn discriminative features from images are the steps. For this medical imaging job, transfer learning algorithms change previously taught models. A collection of photos with established diagnoses is used to test the system’s accuracy, sensitivity, specificity, and performance. The study found that deep learning CNNs can detect oral cancer with 93.62% accuracy and efficiency. High sensitivity and specificity make the trained model a useful early cancer screening tool for healthcare practitioners. These automated methods speed up oral cancer detection and treatment, improving patient outcomes and reducing disease burden.