[1]郝 堃,胡 磊,丁晓明.加权 Fisher 卷积神经网络乳腺癌检测模型[J].计算机技术与发展,2022,32(06):179-185.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 030]
　HAO Kun,HU Lei,DING Xiao-ming.A Model Based on Weighted Fisher LetNet-5 Criterion Convolutional Neural Network for Breast Cancer Detection[J].,2022,32(06):179-185.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 030]
点击复制

加权 Fisher 卷积神经网络乳腺癌检测模型()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年06期
页码:
179-185
栏目:
应用前沿与综合
出版日期:
2022-06-10

文章信息/Info

Title:
A Model Based on Weighted Fisher LetNet-5 Criterion Convolutional Neural Network for Breast Cancer Detection
文章编号:
1673-629X(2022)06-0179-07
作者:
郝 堃; 胡 磊; 丁晓明
1. 重庆医科大学附属第一医院 信息中心,重庆 400010;
2. 西南大学 计算机与信息科学学院,重庆 400010
Author(s):
HAO Kun1 ; HU Lei1 ; DING Xiao-ming2
1. Information Center,the Hospital Group of the First Affiliated Hospital of CQMU,Chongqing 400010,China;
2. School of Computer & Information Science,Southwest University,Chongqing 400010,China
关键词:
乳腺 X 光摄影; 乳腺癌; 医学图像分类; 卷积神经网络; 加权 Fisher 准则
Keywords:
mammography; breast cancer; medical image classification; convolutional neural network; weighted Fisher criterion
分类号:
TP391. 4
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 06. 030
摘要:
乳腺癌是女性发病率最高的癌症,已经严重威胁到女性生命健康。 随着医院影像数据量的爆炸式增长和计算机图像分类技术的不断涌现,基于人工智能技术的医学图像分类也逐渐应用于大型医院的科研、临床应用中。 将这种手段运用于乳腺癌影像检测中,对其辅助诊断和病理研究有着重要的意义。 该文提出一个新的代价函数加权 Fisher 准则并将其应用于乳腺癌检测的卷积神经网络模型中,其目的在于保证图像输出值和样本标签之间的残差最小的同时,使得同类样本距离越近越好,异类样本距离越远越好,从而加强模型对乳腺癌的分类能力。 在公开数据集和医院提取的真实数据上的实验表明,加权 Fisher 准则能有效提升卷积神经网络的收敛时间和识别率,同时基于改进的 LetNet - 5 相较改进的AlexNet 有更优的效果。 将加权 Fisher 的 LetNet-5 卷积神经网络模型用于乳腺癌辅助诊断,具备一定的临床价值和应用前景。
Abstract:
Breast cancer is the cancer with the highest incidence in women and has seriously threatened women’s life and health. With theexplosive growth of health related image data and the continuous emergence of computer image classification technology,medical imageclassification based on artificial intelligence technology has been gradually applied in the scientific research and clinical applications inhospitals. The application of artificial intelligence in breast cancer imaging detection shows great value in auxiliary diagnosis andpathological research. We present a new model for breast cancer detection. In this model,weighted Fisher criterion is used in the LetNet-5 convolutional neural network,which aims to ensure the minimum residual between the output value of image and the sample label,andmake the distance between similar samples as close as possible,and the distance between different samples as far as possible,so as tostrengthen the classification ability of the model for breast cancer. Experiments on public data sets and real data extracted from hospitalsshow that the weighted Fisher criterion can effectively improve the convergence time and recognition rate of convolutional neuralnetworks. At the same time,the improved LetNet-5 is better than the improved AlexNet. The weighted Fisher’s LetNet-5 convolutionalneural network model is used for breast cancer auxiliary diagnosis,which has certain clinical value and application prospects.

相似文献/References:

[1]王兵锐,张新刚,杨晓非.基于精简卷积神经网络的低分辨率乳腺癌识别[J].计算机技术与发展,2020,30(09):200.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 036]
　WANG Bing-rui,ZHANG Xin-gang,YANG Xiao-fei.Low Resolution Breast Cancer Recognition Based on Simplified Convolutional Neural Network[J].,2020,30(06):200.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 036]
[2]韩 笑,毕 波,唐锦萍,等.核零空间方法在乳腺癌异常检测中的应用[J].计算机技术与发展,2022,32(01):165.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 028]
　HAN Xiao,BI Bo,TANG Jin-ping,et al.Application of Kernel Null Space Method in Breast CancerAbnormal Detection[J].,2022,32(06):165.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 028]
[3]张 莉,张 成,郝 岩,等.基于多尺度多色域特征融合的乳腺癌图像分类[J].计算机技术与发展,2022,32(04):176.[doi:10. 3969 / j. issn. 1673-629X. 2022. 04. 030]
　ZHANG Li,ZHANG Cheng,HAO Yan,et al.Breast Cancer Histopathological Image Classification Based on Multi-scale and Multi-gamut Feature Fusion[J].,2022,32(06):176.[doi:10. 3969 / j. issn. 1673-629X. 2022. 04. 030]

更新日期/Last Update: 2022-06-10