Although deep learning based diabetic retinopathy (DR) classification methods typically benefit from well-designed architectures of convolutional neural networks, the training setting also has a non-negligible impact on the prediction performance. The training setting includes various interdependent components, such as objective function, data sampling strategy and data augmentation approach. To identify the key components in a standard deep learning framework (ResNet-50) for DR grading, we systematically analyze the impact of several major components. Extensive experiments are conducted on a publicly-available dataset EyePACS. We demonstrate that (1) the DR grading framework is sensitive to input resolution, objective function, and composition of data augmentation, (2) using mean square error as the loss function can effectively improve the performance with respect to a task-specific evaluation metric, namely the quadratically-weighted Kappa, (3) utilizing eye pairs boosts the performance of DR grading and (4) using data resampling to address the problem of imbalanced data distribution in EyePACS hurts the performance. Based on these observations and an optimal combination of the investigated components, our framework, without any specialized network design, achieves the state-of-the-art result (0.8631 for Kappa) on the EyePACS test set (a total of 42670 fundus images) with only image-level labels. We also examine the proposed training practices on other fundus datasets and other network architectures to evaluate their generalizability. Our codes and pre-trained model are available at https://github.com/YijinHuang/pytorch-classification.
翻译:虽然深层学习基于糖尿病视网膜病(DR)分类方法通常受益于设计完善的神经神经网络结构,但培训设置对预测性业绩也有不可忽略的影响,培训设置包括各种相互依存的组成部分,如客观功能、数据抽样战略和数据增强方法。为了确定标准深层学习框架(ResNet-50)中用于DR分级的关键组成部分(ResNet-50),我们系统地分析几个主要组成部分的影响。在公开可用的数据集 " 眼视系统 " 上进行了广泛的试验。我们表明:(1) DR评级框架对投入分辨率、客观功能和数据增强构成十分敏感,(2)使用中度平方错误,因为损失功能可以有效改进任务特定评价指标方面的业绩,即四重体加权的卡帕方法。(3)利用双眼匹配提高DR评分的性能,(4)利用数据重标来解决在EMEPACS中数据分布不平衡的问题,会损害业绩。基于这些观察和所调查组成部分的最佳组合,我们的框架,没有专门的网络设计,而是使用SBAR-CS-CSlal-al-CSal-Cal-alfal-alfal-Ialalalalalal-Cal-al-al-al-Calal-Calview, 也用我们GServial-CSergal-CSergal-al-al-Supal-Serg-al-al-S-S-aldaldalg-s-S-S-S-laps-al-labalg-s-s-s-sal-salg-salg-lation-al-al-laps,我们总图图图图。我们总图图图图图。