Optimization of Deep Learning Network Parameters Using Uniform Experimental Design for Breast Cancer Histopathological Image Classification.

Cheng-Jian Lin,Shiou-Yun Jeng

doi:10.3390/diagnostics10090662

Abstract

Breast cancer, a common cancer type, is a major health concern in women. Recently, researchers used convolutional neural networks (CNNs) for medical image analysis and demonstrated classification performance for breast cancer diagnosis from within histopathological image datasets. However, the parameter settings of a CNN model are complicated, and using Breast Cancer Histopathological Database data for the classification is time-consuming. To overcome these problems, this study used a uniform experimental design (UED) and optimized the CNN parameters of breast cancer histopathological image classification. In UED, regression analysis was used to optimize the parameters. The experimental results indicated that the proposed method with UED parameter optimization provided 84.41% classification accuracy rate. In conclusion, the proposed method can improve the classification accuracy effectively, with results superior to those of other similar methods.

Highlights

Breast cancer is a commonly diagnosed cancer in women worldwide
The techniques of breast cancer diagnosis depend on investigation of histopathological images such as mammography, magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), thermography, and surgical incision [2,3]
To enhance the classification accuracy, the current study developed a convolution neural networks (CNNs) based on uniform experimental design (UED) to solve the complicated parameter setting problem

Summary

Introduction

Breast cancer is a commonly diagnosed cancer in women worldwide. The accuracy of histopathological image classification is essential for early breast cancer diagnosis. The techniques of breast cancer diagnosis depend on investigation of histopathological images such as mammography, magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), thermography, and surgical incision [2,3]. Advanced engineering techniques have been used by research groups such as the Visual Geometry Group and Google, which have modeled the VGG-16, ResNet and GoogleNet models [6]. These engineering techniques include deep learning models based on convolution neural networks (CNNs), used to improve breast cancer diagnosis efficiency [7]

Objectives

Methods

Results

Conclusion