Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning
Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning
复制标题
DOI：
10.1038/s41746-019-0104-2
复制
发表时间：
2019-04-26
影响因子：
15.2
通讯作者：
Chen, Kuan-Ta
中科院分区：
文献类型：
--
作者：
Kuo, Chin-Chi;Chang, Chun-Min;Chen, Kuan-Ta
长期以来，通过肾脏超声成像预测肾功能和慢性肾脏疾病（CKD）在临床实践中一直被认为是可取的，因为它安全、方便、负担得起。然而，这种非常理想的方法超出了人类视觉的能力。我们开发了一种深度团队方法来自动确定估计的肾小球滤过率（eGFR）和CKD状态。我们利用迁移学习技术，将在ImageNet数据集上预训练的强大ResNet模型集成到我们的神经网络架构中，基于4,505张肾脏超声图像（使用血清肌酐浓度衍生的egfr标记）预测肾功能。为了进一步从超声图像中提取信息，我们利用肾脏长度注释来去除肾脏的外周区域，并应用各种数据增强方案来产生具有变化的附加数据。为了避免过拟合，提高模型的泛化能力，还采用了自举聚合法。此外，我们的深度神经网络获得的肾功能特征被用于识别由eGFR定义的CKD状态
Prediction of kidney function and chronic kidney disease (CKD) through kidney ultrasound imaging has long been considered desirable in clinical practice because of its safety, convenience, and affordability. However, this highly desirable approach is beyond the capability of human vision. We developed a deep teaming approach for automatically determining the estimated glomerular filtration rate (eGFR) and CKD status. We exploited the transfer learning technique, integrating the powerful ResNet model pretrained on an ImageNet dataset in our neural network architecture, to predict kidney function based on 4,505 kidney ultrasound images labeled using eGFRs derived from serum creatinine concentrations. To further extract the information from ultrasound images, we leveraged kidney length annotations to remove the peripheral region of the kidneys and applied various data augmentation schemes to produce additional data with variations. Bootstrap aggregation was also applied to avoid overfitting and improve the model's generalization. Moreover, the kidney function features obtained by our deep neural network were used to identify the CKD status defined by an eGFR of