基于ResNet深度网络的人类蛋白质图谱图像分类方法研究 |
投稿时间:2019-02-19 点此下载全文 |
引用本文:常川.基于ResNet深度网络的人类蛋白质图谱图像分类方法研究[J].医学信息学杂志,2019,40(7):45-49 |
摘要点击次数: |
全文下载次数: |
|
|
中文摘要:将基于深度学习的图像分类方法引入人类蛋白质图谱图像分类中,利用ResNet深度网络构建面向人类蛋白质图谱图像分类的深度卷积神经网络,通过混合模式的蛋白质显微镜图像进行验证。结果表明该方法比其他自动分类法具有更高的准确率和精度,大大节约人力和时间。 |
中文关键词:深度学习 ResNet网络 医学影像 图像分类 |
|
Study of Image Classification Method of Human Protein Atlas Based on ResNet Deep Network |
|
|
Abstract:The paper introduces the image classification method based on deep learning into the image classification method of human protein atlas. It uses ResNet deep network to build a deep convolutional neural network for the image classification of human protein atlas, and verifies by using the microscope image of mixed-mode protein. The result shows that this method has higher accuracy and precision than other automatic classification methods, and greatly saves time and manpower. |
keywords:deep learning ResNet network medical image image classification |
查看全文 查看/发表评论 下载PDF阅读器 |