基于卷积神经网络的动态对比增强磁共振成像中的自动肾脏分割
AUTOMATIC RENAL SEGMENTATION IN DCE-MRI USING CONVOLUTIONAL NEURAL NETWORKS.
作者信息
Haghighi Marzieh, Warfield Simon K, Kurugol Sila
机构信息
Electrical and Computer Engineering Department, Northeastern University, Boston, MA, 02115.
Radiology Department, Boston Childrens Hospital; and Harvard Medical School, Boston MA 02115.
出版信息
Proc IEEE Int Symp Biomed Imaging. 2018 Apr;2018:1534-1537. doi: 10.1109/ISBI.2018.8363865. Epub 2018 May 24.
Abstract
Kidney function evaluation using dynamic contrast-enhanced MRI (DCE-MRI) images could help in diagnosis and treatment of kidney diseases of children. Automatic segmentation of renal parenchyma is an important step in this process. In this paper, we propose a time and memory efficient fully automated segmentation method which achieves high segmentation accuracy with running time in the order of seconds in both normal kidneys and kidneys with hydronephrosis. The proposed method is based on a cascaded application of two 3D convolutional neural networks that employs spatial and temporal information at the same time in order to learn the tasks of localization and segmentation of kidneys, respectively. Segmentation performance is evaluated on both normal and abnormal kidneys with varying levels of hydronephrosis. We achieved a mean dice coefficient of 91.4 and 83.6 for normal and abnormal kidneys of pediatric patients, respectively.
摘要
使用动态对比增强磁共振成像（DCE-MRI）图像进行肾功能评估有助于儿童肾脏疾病的诊断和治疗。肾实质的自动分割是这一过程中的重要步骤。在本文中，我们提出了一种高效省时且节省内存的全自动分割方法，该方法在正常肾脏和肾积水肾脏中均能以秒级的运行时间实现较高的分割精度。所提出的方法基于两个3D卷积神经网络的级联应用，这两个网络同时利用空间和时间信息，分别学习肾脏的定位和分割任务。在具有不同肾积水程度的正常和异常肾脏上评估分割性能。我们分别在儿科患者的正常和异常肾脏上实现了平均骰子系数为91.4和83.6。
相似文献
1
AUTOMATIC RENAL SEGMENTATION IN DCE-MRI USING CONVOLUTIONAL NEURAL NETWORKS.基于卷积神经网络的动态对比增强磁共振成像中的自动肾脏分割
Proc IEEE Int Symp Biomed Imaging. 2018 Apr;2018:1534-1537. doi: 10.1109/ISBI.2018.8363865. Epub 2018 May 24.
2
Improving Automatic Renal Segmentation in Clinically Normal and Abnormal Paediatric DCE-MRI via Contrast Maximisation and Convolutional Networks for Computing Markers of Kidney Function.通过对比度最大化和卷积网络提高临床正常和异常儿科 DCE-MRI 的自动肾脏分割，以计算肾功能标志物。
Sensors (Basel). 2021 Nov 28;21(23):7942. doi: 10.3390/s21237942.
3
Healthy Kidney Segmentation in the Dce-Mr Images Using a Convolutional Neural Network and Temporal Signal Characteristics.使用卷积神经网络和时间信号特征对 DCE-MR 图像进行健康肾脏分割。
Sensors (Basel). 2021 Oct 9;21(20):6714. doi: 10.3390/s21206714.
4
Kidney Segmentation from Dynamic Contrast-Enhanced Magnetic Resonance Imaging Integrating Deep Convolutional Neural Networks and Level Set Methods.基于深度卷积神经网络与水平集方法融合的动态对比增强磁共振成像肾脏分割
Bioengineering (Basel). 2023 Jun 24;10(7):755. doi: 10.3390/bioengineering10070755.
5
A self-supervised strategy for fully automatic segmentation of renal dynamic contrast-enhanced magnetic resonance images.一种用于肾动态对比增强磁共振图像全自动分割的自监督策略。
Med Phys. 2019 Oct;46(10):4417-4430. doi: 10.1002/mp.13715. Epub 2019 Aug 16.
6
A novel MRI segmentation method using CNN-based correction network for MRI-guided adaptive radiotherapy.一种基于 CNN 的修正网络的新型 MRI 分割方法，用于 MRI 引导自适应放疗。
Med Phys. 2018 Nov;45(11):5129-5137. doi: 10.1002/mp.13221. Epub 2018 Oct 28.
7
Automatic abdominal multi-organ segmentation using deep convolutional neural network and time-implicit level sets.基于深度卷积神经网络和时间隐式水平集的自动腹部多器官分割。
Int J Comput Assist Radiol Surg. 2017 Mar;12(3):399-411. doi: 10.1007/s11548-016-1501-5. Epub 2016 Nov 24.
8
Spatial aggregation of holistically-nested convolutional neural networks for automated pancreas localization and segmentation.整体嵌套卷积神经网络的空间聚合用于自动胰腺定位和分割。
Med Image Anal. 2018 Apr;45:94-107. doi: 10.1016/j.media.2018.01.006. Epub 2018 Feb 1.
9
Harnessing Artificial Intelligence for Enhanced Renal Analysis: Automated Detection of Hydronephrosis and Precise Kidney Segmentation.利用人工智能增强肾脏分析：肾积水的自动检测与精确肾脏分割
Eur Urol Open Sci. 2024 Feb 22;62:19-25. doi: 10.1016/j.euros.2024.01.017. eCollection 2024 Apr.
10
Multiregion segmentation of bladder cancer structures in MRI with progressive dilated convolutional networks.MRI 中基于渐进式扩张卷积网络的膀胱癌结构的多区域分割。
Med Phys. 2018 Dec;45(12):5482-5493. doi: 10.1002/mp.13240. Epub 2018 Nov 8.
引用本文的文献
1
CNN-Based Kidney Segmentation Using a Modified CLAHE Algorithm.基于卷积神经网络的肾脏分割：使用改进的对比度受限自适应直方图均衡化算法
Sensors (Basel). 2024 Dec 2;24(23):7703. doi: 10.3390/s24237703.
2
Feature Extraction Based on Local Histogram with Unequal Bins and a Recurrent Neural Network for the Diagnosis of Kidney Diseases from CT Images.基于不等距区间局部直方图和循环神经网络的CT图像肾脏疾病诊断特征提取
Bioengineering (Basel). 2024 Feb 25;11(3):220. doi: 10.3390/bioengineering11030220.
3
Kidney Segmentation from Dynamic Contrast-Enhanced Magnetic Resonance Imaging Integrating Deep Convolutional Neural Networks and Level Set Methods.基于深度卷积神经网络与水平集方法融合的动态对比增强磁共振成像肾脏分割
Bioengineering (Basel). 2023 Jun 24;10(7):755. doi: 10.3390/bioengineering10070755.
4
In vivo MRI is sensitive to remyelination in a nonhuman primate model of multiple sclerosis.体内 MRI 对多发性硬化的非人灵长类动物模型中的髓鞘再生敏感。
Elife. 2023 Apr 21;12:e73786. doi: 10.7554/eLife.73786.
5
Deep-learning-based ensemble method for fully automated detection of renal masses on magnetic resonance images.基于深度学习的集成方法用于磁共振图像上肾肿块的全自动检测
J Med Imaging (Bellingham). 2023 Mar;10(2):024501. doi: 10.1117/1.JMI.10.2.024501. Epub 2023 Mar 20.
6
Validation of automatically measured T1 map cortico-medullary difference (ΔT1) for eGFR and fibrosis assessment in allograft kidneys.验证自动测量的皮质-髓质 T1 映射差值 (ΔT1) 用于评估同种异体移植肾脏的 eGFR 和纤维化。
PLoS One. 2023 Feb 15;18(2):e0277277. doi: 10.1371/journal.pone.0277277. eCollection 2023.
7
Variational Approach for Joint Kidney Segmentation and Registration from DCE-MRI Using Fuzzy Clustering with Shape Priors.基于模糊聚类与形状先验的DCE-MRI联合肾脏分割与配准的变分方法
Biomedicines. 2022 Dec 21;11(1):6. doi: 10.3390/biomedicines11010006.
8
Level-Set-Based Kidney Segmentation from DCE-MRI Using Fuzzy Clustering with Population-Based and Subject-Specific Shape Statistics.基于水平集的动态对比增强磁共振成像肾脏分割：使用基于总体和个体形状统计的模糊聚类方法
Bioengineering (Basel). 2022 Nov 5;9(11):654. doi: 10.3390/bioengineering9110654.
9
Kidney segmentation from DCE-MRI converging level set methods, fuzzy clustering and Markov random field modeling.基于 DCE-MRI 的水平集融合方法、模糊聚类和马尔可夫随机场模型的肾脏分割。
Sci Rep. 2022 Nov 5;12(1):18816. doi: 10.1038/s41598-022-23408-1.
10
Prospective motion correction in kidney MRI using FID navigators.前瞻性运动校正在肾 MRI 中的应用。
Magn Reson Med. 2023 Jan;89(1):276-285. doi: 10.1002/mrm.29424. Epub 2022 Sep 5.
本文引用的文献
1
Auto-Context Convolutional Neural Network (Auto-Net) for Brain Extraction in Magnetic Resonance Imaging.用于磁共振成像中脑提取的自动上下文卷积神经网络（自动网络）
IEEE Trans Med Imaging. 2017 Nov;36(11):2319-2330. doi: 10.1109/TMI.2017.2721362. Epub 2017 Jun 28.
2
Automatic renal segmentation for MR urography using 3D-GrabCut and random forests.基于 3D-GrabCut 和随机森林的磁共振尿路成像自动肾脏分割。
Magn Reson Med. 2018 Mar;79(3):1696-1707. doi: 10.1002/mrm.26806. Epub 2017 Jun 27.
3
Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation.高效多尺度 3D CNN 结合全连接条件随机场实现精准脑损伤分割。
Med Image Anal. 2017 Feb;36:61-78. doi: 10.1016/j.media.2016.10.004. Epub 2016 Oct 29.
4
Renal compartment segmentation in DCE-MRI images.DCE-MRI 图像的肾实质分割。
Med Image Anal. 2016 Aug;32:269-80. doi: 10.1016/j.media.2016.05.006. Epub 2016 May 16.
5
Golden-angle radial sparse parallel MRI: combination of compressed sensing, parallel imaging, and golden-angle radial sampling for fast and flexible dynamic volumetric MRI.黄金角径向稀疏并行磁共振成像：压缩感知、并行成像和黄金角径向采样相结合实现快速灵活的动态容积磁共振成像。
Magn Reson Med. 2014 Sep;72(3):707-17. doi: 10.1002/mrm.24980. Epub 2013 Oct 18.
6
Functional analysis in MR urography - made simple.磁共振尿路成像中的功能分析——简明扼要。
Pediatr Radiol. 2010 Feb;40(2):182-99. doi: 10.1007/s00247-009-1458-4. Epub 2009 Dec 12.
7
Assessment of 3D DCE-MRI of the kidneys using non-rigid image registration and segmentation of voxel time courses.使用非刚性图像配准和体素时间历程分割对肾脏进行三维动态对比增强磁共振成像评估。
Comput Med Imaging Graph. 2009 Apr;33(3):171-81. doi: 10.1016/j.compmedimag.2008.11.004. Epub 2009 Jan 9.
Zotero 插件