Comparing different deep learning architectures for classification of chest radiographs
Comparing different deep learning architectures for classification of chest radiographs
复制标题
DOI：
10.1038/s41598-020-70479-z
复制
发表时间：
2020-08-12
影响因子：
4.6
通讯作者：
Vahldiek, Janis L.
中科院分区：
文献类型：
--
作者：
Bressem, Keno K.;Adams, Lisa C.;Vahldiek, Janis L.
胸部X光片是放射学中最常获得的图像之一，也是计算机视觉研究的对象。然而，大多数用于胸片分类的模型是从开放可用的深度神经网络派生出来的，这些神经网络是在大型图像数据集上训练的。这些数据集与胸部X光照片的不同之处在于，它们大多是彩色图像，并且具有更多的标签。因此，对于相对简单的医学图像数据分类任务，可能不需要为ImageNet设计的非常深的卷积神经网络(CNN)，并且通常表示更复杂的关系。比较了16种不同结构的CNN在两个公开可用的数据集CheXpert和新冠肺炎图像数据集合上的分类性能。在CheXpert数据集上可以获得在0.83到0.89之间的接收器工作特性曲线(AUROC)下的区域。在新冠肺炎图像数据采集上，所有模型都显示出出色的能力，可以检测出AUROC值在0.983到0.998之间的新冠肺炎和非冠状病毒肺炎。可以观察到，更浅的网络可能以更短的训练时间获得与更深和更复杂的对应网络相当的结果，使得即使在使用有限的硬件时，对医学图像数据的分类性能也接近最先进的方法。
Chest radiographs are among the most frequently acquired images in radiology and are often the subject of computer vision research. However, most of the models used to classify chest radiographs are derived from openly available deep neural networks, trained on large image datasets. These datasets differ from chest radiographs in that they are mostly color images and have substantially more labels. Therefore, very deep convolutional neural networks (CNN) designed for ImageNet and often representing more complex relationships, might not be required for the comparably simpler task of classifying medical image data. Sixteen different architectures of CNN were compared regarding the classification performance on two openly available datasets, the CheXpert and COVID-19 Image Data Collection. Areas under the receiver operating characteristics curves (AUROC) between 0.83 and 0.89 could be achieved on the CheXpert dataset. On the COVID-19 Image Data Collection, all models showed an excellent ability to detect COVID-19 and non-COVID pneumonia with AUROC values between 0.983 and 0.998. It could be observed, that more shallow networks may achieve results comparable to their deeper and more complex counterparts with shorter training times, enabling classification performances on medical image data close to the state-of-the-art methods even when using limited hardware.