Harnessing deep-learning techniques for early prognosis of oral cancer detection

DOI

10.2991/978-94-6463-738-0_15How to use a DOI?

Keywords

Oral Squamous Cell Carcinoma; CANet; Ensemble model; EfficientNetB3; ResNet50

Abstract

The global burden of oral cancer continues to be a major challenge in the field of health, which demands the development and establishment of efficient diagnostic methods. This study introduces a novel ensemble model that harnesses the benefits of two best-performing convolutional neural networks in its design; EfficientNetB3 and ResNet50. The proposed architecture optimizes filter computation and enhances the capacity of extracting features through transfer learning with selective layer freezing, minimizing the overfitting process using strategic dropout. This model was developed using a large dataset of oral cancer images and employed an aggressive data augmentation strategy to enhance generalization. Accuracy was computed on the validation dataset and training dataset with benign vs malignant labels which shows that lesions can be effectively classified as malignant or benign by the model. Collectively the ensemble approach likewise statistically dominated individual models, further supporting it to be a dependable detection tool for early oral cancer diagnosis. These findings add to the ongoing work in developing more accurate diagnostic tools for oncology and highlight that using advanced machine-learning approaches can be beneficial while dealing with medical imaging.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Bhargavi Devapatla
AU  - Manikanta Kancharla
AU  - Mamta Arora
PY  - 2025
DA  - 2025/06/22
TI  - Harnessing deep-learning techniques for early prognosis of oral cancer detection
BT  - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025)
PB  - Atlantis Press
SP  - 179
EP  - 191
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-738-0_15
DO  - 10.2991/978-94-6463-738-0_15
ID  - Devapatla2025
ER  -