Pre-training on Grayscale ImageNet Improves Medical Image Classification
Paper:http://openaccess.thecvf.com/content_ECCVW_2018/papers/11134/Xie_Pre-training_on_Grayscale_ImageNet_Improves_Medical_Image_Classification_ECCVW_2018_paper.pdf
Tips:ECCV2018的一篇paper。
(阅读笔记)
1. Main idea
- 提出问题,医学图像的获取很困难。large medical image datasets appropriate for training deep neural network models from scratch are difficult to assemble due to privacy restrictions and expert ground truth requirements.
- 一般的解决方法即是通过预训练一个其他的模型再微调。to pre-train models on large datasets in other domains, such as ImageNet for classification of natural images.
- 再一次提出问题,Image-Net的图像都是三通道,而医学图像一般都是灰度单通道图像。However, ImageNet contains color images, which introduces artefacts and inefficiencies into models that are intended for single-channel medical images.
- 所以提出了本文的idea。we pre-trained an Inception-V3 model on ImageNet after converting the images to grayscale through a common transformation.
2. Method
把Image-Net的图像转换成灰度图像后,用相同的参数喂入网络(仅改变通道数为1),得到预训练的模型,然后微调后实验效果很好。
![在这里插入图片描述](https://img-blog.csdnimg.cn/20200316210824101.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQyMTkyOTEw,size_16,color_FFFFFF,t_70)