期刊文献+

深度学习技术在乳腺癌诊断中的应用 认领

Application of deep learning technology in breast cancer diagnosis
在线阅读 下载PDF
收藏 分享 导出
摘要 目的构建深度学习卷积神经网络模型,提高乳腺癌诊断的智能化和信息化水平。方法通过对真实临床中公开的乳腺癌数据集进行统计分析,运用人工智能领域的卷积神经网络模型,为医疗人员诊断恶性乳腺癌患者提供可靠的理论基础。根据建立的神经网络模型,选取当前流行的乳腺癌数据集进行建模分析,得到相应的诊断结果。结果实验结果显示,大数据驱动下的乳腺癌诊断模型能够准确有效预测恶性乳腺癌患者。该文提出的卷积神经网络模型与传统的支持向量机模型相比,在准确率、特异性、敏感性和曲线下面积方面相比分别提高2.7%、2.9%、2.8%和3.0%。结论深度学习领域的卷积神经网络方法,在乳腺癌诊断方面具有良好的前景,可减少医疗人员的病情诊断负担,为人工智能视角下的乳腺癌诊断技术指明新方向。 【Objective】In order to improve the intelligence and information level of breast cancer diagnosis,a deep learning convolution neural network model is constructed.【Methods】Through the statistical analysis of the real open breast cancer data set,the convolution neural network model in the field of artificial intelligence is used to provide a reliable theoretical basis for medical staff to diagnose the patients with malignant breast cancer.According to the established neural network model,the breast cancer data set for modeling and analysis was selected,and the corresponding diagnosis results were obtained.【Results】The experimental results show that the large data driven diagnosis model of breast cancer can accurately and effectively predict the patients with malignant breast cancer.Compared with the traditional support vector machine model,the convolution neural network model proposed in this paper improves the accuracy,specificity,sensitivity and AUC by 2.7%,2.9%,2.8%and 3.0%respectively.【Conclusion】The convolution neural network method in the field of degree learning has a good prospect in the field of breast cancer diagnosis,which greatly reduces the burden of disease diagnosis of medical personnel,and points out a new direction for breast cancer diagnosis technology from the perspective of artificial intelligence.
作者 李蒙蒙 LI Mengmeng(The First Affiliated Hospital of Henan University of Science and Technology,LuoYang,Henan 471000,China)
出处 《中国医学工程》 2021年第1期1-3,共3页 China Medical Engineering
关键词 乳腺癌 人工智能 计算机辅助诊断 深度学习 breast cancer artificial intelligence computer-aided diagnosis deep learning
  • 相关文献

参考文献4

二级参考文献17

共引文献44

202103读书月活动
维普数据出版直通车
今日学术
投稿分析
职称考试

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部 意见反馈