Deep Learning on Histopathological Images: Automated Classification of Oral Squamous Cell Carcinoma Stages Detection using Pre-trained Convolutional Neural Networks
P. Aurchana1, P. Dhanalakshmi2

1P. Aurchana, Research Scholar, Professor, Department of CSE, Annamalai University, Tamil Nadu, India.
2P. Dhanalakshmi, Professor, Department of CSE, Annamalai University, Tamil Nadu, India.
Manuscript received on December 02, 2020 | Revised Manuscript Received on  12, 2020. | Manuscript published on December 10, 2020. | PP: 17-23 | Volume-1 Issue-1, December 2020 |  Retrieval Number: A1004061120/2020©LSP
Open Access | Ethics and Policies
© The Authors. Published by Lattice Science Publication (LSP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Deep neural networks have gained immense popularity in the recent years by achieving very good results in medical analysis. This work aims at finding the stages of Oral squamous cell carcinoma using the Pre-trained convolution neural network models like Alexnet, Googlenet and Resnet50. Every pathologist while evaluating the photomicrograph finds it difficult in analyzing the stages of oral squamous cell carcinoma into poorly differentiated, moderately differentiated and well differentiated. To overcome, this deep convolution neural network model has been implemented. In the proposed work, Deep learning needs huge amount of data to achieve good performance so image augmentation has been implemented to boost the performance in deep networks. Later segmentation has been implemented and it is given to the Pre-trained convolutional neural networks which gives the satisfactory result of above80 %.GoogleNet gives the highest results than Alexnet and Resnet50.
Keywords: Oral Squamous Cell Carcinoma (OSCC), Convention Neural Network (CNN), Alexnet, Google Net, Resnet50.