Classification of Benign and Malignant Breast Cancer Diagnosis Based on Convolutional Neural Networks
- Resource Type
- Conference
- Authors
- Gao, Juan; Qing, Xiaoyan; Li, Mingjin; Xie, Shuang; Xiang, Qin; Cai, Weijie
- Source
- 2024 IEEE 7th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC) Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), 2024 IEEE 7th. 7:1231-1236 Mar, 2024
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Robotics and Control Systems
Signal Processing and Analysis
Training
Pathology
Convolution
Computational modeling
Feature extraction
Breast cancer
Data models
Pathological images of breast cancer
image classification
Convolutional neural networks
- Language
- ISSN
- 2689-6621
Efficient and accurate classification of breast cancer pathological images is of crucial significance for both doctors’ diagnosis and computer-aided diagnosis, and this study combines the problems existing in the classification of breast cancer pathological images at this stage, and proposes a method based on convolutional neural network (VGG_16) to extract features of breast cancer pathological images to realize the pathological images of breast cancer, which is faster in the processing of breast cancer pathological images and can quickly and accurately diagnose benign and malignant lesions. By learning from different details and textures in the image, the network is able to effectively capture important features about benign and malignant breast cancer. The continuously optimized model can achieve a classification accuracy of 98% or more for benign and malignant breast cancer on the BreakHis dataset. It aims to improve the accuracy and efficiency of benign and malignant diagnosis of breast cancer. Experimental results show that the classification method for benign and malignant diagnosis of breast cancer based on convolutional neural network is effective. This has important implications for both diagnostic decision-making by doctors and computer-aided diagnosis. Further research can explore deeper network structures or other improved methods to further improve the performance and accuracy of breast cancer pathological image classification.