Tang Qiling, Liu Yangyang, Liu Haihua
College of Biomedical Engineering, South Central University for Nationalities, Wuhan 430074, PR China.
Huibei Key Laboratory for Medical Information Analysis and Tumor Treatment, Wuhan 430074, PR China.
Artif Intell Med. 2017 Jun;79:71-78. doi: 10.1016/j.artmed.2017.06.009. Epub 2017 Jun 29.
Multiscale structure is an essential attribute of natural images. Similarly, there exist scaling phenomena in medical images, and therefore a wide range of observation scales would be useful for medical imaging measurements. The present work proposes a multiscale representation learning method via sparse autoencoder networks to capture the intrinsic scales in medical images for the classification task. We obtain the multiscale feature detectors by the sparse autoencoders with different receptive field sizes, and then generate the feature maps by the convolution operation. This strategy can better characterize various size structures in medical imaging than single-scale version. Subsequently, Fisher vector technique is used to encode the extracted features to implement a fixed-length image representation, which provides more abundant information of high-order statistics and enhances the descriptiveness and discriminative ability of feature representation. We carry out experiments on the IRMA-2009 medical collection and the mammographic patch dataset. The extensive experimental results demonstrate that the proposed method have superior performance.
多尺度结构是自然图像的一个基本属性。同样，医学图像中也存在尺度现象，因此广泛的观察尺度对于医学成像测量是有用的。目前的工作提出了一种通过稀疏自动编码器网络的多尺度表示学习方法，以捕捉医学图像中的内在尺度用于分类任务。我们通过具有不同感受野大小的稀疏自动编码器获得多尺度特征检测器，然后通过卷积操作生成特征图。与单尺度版本相比，这种策略可以更好地表征医学成像中的各种大小结构。随后，使用Fisher向量技术对提取的特征进行编码，以实现固定长度的图像表示，这提供了更丰富的高阶统计信息，并增强了特征表示的描述性和判别能力。我们在IRMA - 2009医学数据集和乳腺X线摄影补丁数据集上进行了实验。广泛的实验结果表明，所提出的方法具有优异的性能。
