Automated renal segmentation in healthy and chronic kidney disease subjects using a convolutional neural network
Automated renal segmentation in healthy and chronic kidney disease subjects using a convolutional neural network
复制标题
DOI：
10.1002/mrm.28768
复制
发表时间：
2021-03-23
影响因子：
3.3
通讯作者：
Francis, Susan T.
中科院分区：
文献类型：
--
作者：
Daniel, Alexander J.;Buchanan, Charlotte E.;Francis, Susan T.
目的：肾脏总体积（TKV）是肾脏疾病检测和监测的重要指标。我们开发了一种完全自动化的方法，从T-2加权MRI分割肾脏，以计算健康对照（HC）和慢性肾脏病（CKD）patients.Methods的TKV：这种自动化方法使用机器学习，特别是2D卷积神经网络（CNN），从T-2加权MRI数据准确分割左右肾。数据集包括30例HC受试者和30例CKD患者。该模型在50个手动定义的HC和CKD肾脏分割上进行训练。该模型随后在50个测试数据集上进行了评估，包括来自5个HC和5个CKD患者的数据，每个患者在扫描会话中扫描5次，以比较CNN和手动分割肾脏的精度。手动和自动计算的TKV之间的差异为1.2 +/- 16.2 mL，平均表面距离为0.65 +/- 0.21 mm。与手动肾脏分割相比，使用自动方法对同一受试者重复采集的TKV测量值的方差显著更低。2D CNN方法提供了左肾和右肾的全自动分割以及TKV的计算，
Purpose: Total kidney volume (TKV) is an important measure in renal disease detection and monitoring. We developed a fully automated method to segment the kidneys from T-2-weighted MRI to calculate TKV of healthy control (HC) and chronic kidney disease (CKD) patients.Methods: This automated method uses machine learning, specifically a 2D convolutional neural network (CNN), to accurately segment the left and right kidneys from T-2-weighted MRI data. The data set consisted of 30 HC subjects and 30 CKD patients. The model was trained on 50 manually defined HC and CKD kidney segmentations. The model was subsequently evaluated on 50 test data sets, comprising data from 5 HCs and 5 CKD patients each scanned 5 times in a scan session to enable comparison of the precision of the CNN and manual segmentation of kidneys.Results: The unseen test data processed by the 2D CNN had a mean Dice score of 0.93 +/- 0.01. The difference between manual and automatically computed TKV was 1.2 +/- 16.2 mL with a mean surface distance of 0.65 +/- 0.21 mm. The variance in TKV measurements from repeat acquisitions on the same subject was significantly lower using the automated method compared to manual segmentation of the kidneys.Conclusion: The 2D CNN method provides fully automated segmentation of the left and right kidney and calculation of TKV in