A Deep Convolutional Neural Network Architecture for Cancer Diagnosis using Histopathological Images

Karthika Gidijala,Manasa Vavilapalli,Mansa Devi Pappu,Mahesh Kothuru,

doi:10.35940/ijitee.l9524.10101221

Karthika Gidijala, Manasa Vavilapalli + Show 3 more

Open Access

https://doi.org/10.35940/ijitee.l9524.10101221

Copy DOI

Abstract

Many different models of Convolution Neural Networks exist in the Deep Learning studies. The application and prudence of the algorithms is known only when they are implemented with strong datasets. The histopathological images of breast cancer are considered as to have much number of haphazard structures and textures. Dealing with such images is a challenging issue in deep learning. Working on wet labs and in coherence to the results many research have blogged with novel annotations in the research. In this paper, we are presenting a model that can work efficiently on the raw images with different resolutions and alleviating with the problems of the presence of the structures and textures. The proposed model achieves considerably good results useful for decision making in cancer diagnosis.

Highlights

The assuages on saving patients life, the research experts rely on sophisticated Computer Aided Diagnosis (CAD) systems
A Deep CNN architecture is implemented to unveil the important aspects of the malignity related pathological symptoms
In order to process the results of all of the pre-configured layers, including the activation results from ReLU, with 240 filters, a fifth convolution layer with 240 filters is fed the activation from ReLU, and a kernel size of 12 × 12 × 240 is used in order to accept all of the results from previously configured layers and maximise the pooling from that layer with stride of 2 × 2

Summary

Introduction

The massive growth of breast cancer and its incidence in the recent years, death rate has been considerably reduced in most developed countries. Many breakthroughs are notified in early detection and screening methods through medical imaging and analysis. Classification of patients with cancer or no-cancer using clinical records requires careful scrutiny with high sensitivity and specificity in a diagnostic test. The core of the consensus in breast cancer research deals with study of multiple magnification levels of histopathological images, which is time-hard and paves into multiple theses of analyses. The assuages on saving patients life, the research experts rely on sophisticated Computer Aided Diagnosis (CAD) systems. The in vitro golden source of datasets are collected in the process for decision making with factorized right time, right diagnosis

Objectives

Results

Conclusion