Automated Oral Cancer Identification: Deep Learning Integration with AI-Enhanced Diagnostic Pathways
[1]Amulya S S, [2]Dr. Nanditha B R, [3]H M Chinthana, [4]Amulya Patel, [5]Gowri S P
[1]Information and Science and Engineering, Malnad College of Engineering Hassan-573202, India
[2]Associate Professor, Information and Science and Engineering, Malnad College of Engineering Hassan-573202, India [3]Information and Science and Engineering, Malnad College of Engineering Hassan-573202, India [4]Information and Science and Engineering, Malnad College of Engineering Hassan-573202, India [5]Information and Science and Engineering, Malnad College of Engineering Hassan-573202, India
Email id: amulyasrinivas86@gmail.com, brn@mcehassan.ac.in, hmchinthana21@gmail.com ,amulyapatel183@gmail.com, gowri03sp@gmail.com
Abstract— Oral cavity cancer is a growing health concern, particularly in areas where the consumption of tobacco and other harmful substances is high. This disease is often diagnosed at a later stage, making treatment more difficult and increasing the mortality rate. Artificial intelligence, especially deep learning-based image classification, offers an effective approach to identifying cancerous regions at an early stage. Therefore, our study proposes to analyze and implement AI techniques to classify intraoral images into two categories: Cancerous or Non- cancerous. Early detection allows for timely medical intervention, reducing complications and improving survival outcomes. In this research, we developed and trained deep learning models using Convolutional Neural Networks (CNNs), focusing on binary classification. We compared the performance of two architectures — VGG19 and Dense Net — on a curated dataset of oral images. Among these, the VGG19 model showed superior accuracy in prediction. The project demonstrates how AI and deep learning can be leveraged to build low-cost, scalable, and accurate diagnostic tools that can potentially save lives by enabling timely intervention. With further enhancement and clinical validation, this model can be deployed in hospitals, dental clinics, and mobile health units for mass screening of oral cavity cancer.
Keywords— Machine learning, Deep learning, Oral Cavity Cancer diagnosis, Image processing, Convolutional neural networks, Medical imaging.
