Integrating Handcrafted and Deep Features for Optical Coherence Tomography Based Retinal Disease Classification
Integrating Handcrafted and Deep Features for Optical Coherence Tomography Based Retinal Disease Classification
复制标题
集成手工制作和深度特征，用于基于光学相干断层扫描的视网膜疾病分类
DOI：
10.1109/access.2019.2891975
复制
发表时间：
2019-01-01
期刊：
影响因子：
3.9
通讯作者：
Qiu, Connor S.
中科院分区：
文献类型：
--
作者：
Li, Xuechen;Shen, Linlin;Qiu, Connor S.
深度神经网络（DNN）已广泛应用于疾病诊断的医学图像自动分析，并通过有效处理大量图像来帮助人类专家。虽然自20世纪90年代以来，手工制作的功能一直用于眼部疾病检测或分类，但DNN最近在这一领域被采用，并表现出非常有前途的性能。由于手工和深度特征可以提取互补信息，因此本文提出了三种不同的集成框架，将联合收割机手工和深度特征结合起来，用于基于光学相干断层扫描图像的眼病分类。此外，为了使用现有的网络（如VGG，DenseNet和Xception）在输入层和全连接层集成手工制作的特征，还提出了一种用于中间层特征集成的新型胸腔网络（RC Net）。对于RC Net，两个“肋骨”通道被设计用于独立处理深度和手工特征，另一个所谓的“脊柱”通道被设计用于集成。而密集块是三个通道的主要组成部分，求和运算被提出用于特征图集成。我们的实验结果表明，深度网络在集成了手工特征后实现了更好的分类准确性，例如，尺度不变特征变换和Gabor.在所有提出的特征集成方法中，RC网表现出最好的性能。
Deep neural networks (DNNs) have been widely applied to the automatic analysis of medical images for disease diagnosis and to help human experts by efficiently processing immense amounts of images. While the handcrafted feature has been used for eye disease detection or classification since the 1990s, DNN was recently adopted in this area and showed a very promising performance. Since handcrafted and deep feature can extract complementary information, we propose, in this paper, three different integration frameworks to combine handcrafted and deep feature for optical coherence tomography image-based eye disease classification. In addition, to integrate the handcrafted feature at the input and fully connected layers using existing networks, such as VGG, DenseNet, and Xception, a novel ribcage network (RC Net) is also proposed for feature integration at middle layers. For RC Net, two “rib” channels are designed to independently process deep and handcrafted features, and another so-called “spine” channel is designed for the integration. While dense blocks are the main components of the three channels, sum operation is proposed for the feature map integration. Our experimental results showed that the deep networks achieved better classification accuracy after the integration of the handcrafted features, e.g., scale-invariant feature transform and Gabor. The RC Net showed the best performance among all the proposed feature integration methods.