摘要
利用威斯康星乳腺癌数据集对乳腺癌智能辅助诊断开展了深入的研究和分析,给出了一种基于神经网络的乳腺癌辅助诊断分析方法.首先对数据进行归一化预处理,然后基于Tensorflow深度学习框架搭建了神经网络模型并进行辅助诊断模型的训练和评估,并进一步采用交叉验证方法对模型稳定性进行了验证,与其他传统机器方法进行对比分析.实验结果表明,给出的方法对于乳腺癌辅助诊断具有较高的准确率和良好的稳定性,为乳腺癌诊断精度的提升提供了智能化方法,对乳腺癌的临床诊断和治疗具有重要的现实意义.
The Wisconsin breast cancer dataset is used to conduct in-depth research and analysis on breast cancer intelligent assisted diagnosis,and a neural network-based breast cancer assisted diagnosis analysis method is presented.Firstly,the data is normalized preprocessing.Then the neural network model is built based on the Tensorflow deep learning framework and the training and evaluation of the auxiliary diagnosis model is carried out.Furthermore,cross-validation of method is used to verify the stability of the model.Finally,it is compared with other traditional machine methods.The experimental results show that the proposed method has high accuracy and stability for breast cancer diagnosis,and provides an intelligent method for the improvement of the diagnosis accuracy of breast cancer,it has important practical significance for the clinical diagnosis and treatment of breast cancer.
作者
张剑飞
崔文升
刘明
杜晓昕
ZHANG Jian-fei;CUI Wen-sheng;LIU Ming;DU Xiao-xin(School of Computer and Control Engineering,Qiqihar University,Qiqihar 161006,China)
出处
《高师理科学刊》
2019年第5期21-25,29,共6页
Journal of Science of Teachers＇College and University
基金
黑龙江省教育厅基本科研业务费科研项目(135309471)
关键词
乳腺癌
神经网络
交叉验证
breast cancer
neural network
cross validation
