用于脑磁共振成像中小病变分割的全卷积神经网络
ORCHESTRAL FULLY CONVOLUTIONAL NETWORKS FOR SMALL LESION SEGMENTATION IN BRAIN MRI.
作者信息
Xu Botian, Chai Yaqiong, Galarza Cristina M, Vu Chau Q, Tamrazi Benita, Gaonkar Bilwaj, Macyszyn Luke, Coates Thomas D, Lepore Natasha, Wood John C
机构信息
CIBORG laboratory, Department of Radiology, Children's Hospital Los Angeles (CHLA).
Department of Electrical Engineering, USC.
出版信息
Proc IEEE Int Symp Biomed Imaging. 2018 Apr;2018:889-892. doi: 10.1109/ISBI.2018.8363714. Epub 2018 May 24.
Abstract
White matter (WM) lesion identification and segmentation has proved of clinical importance for diagnosis, treatment and neurological outcomes. Convolutional neural networks (CNN) have demonstrated their success for large lesion load segmentation, but are not sensitive to small deep WM and sub-cortical lesion segmentation. We propose to use multi-scale and supervised fully convolutional networks (FCN) to segment small WM lesions in 22 anemic patients. The multiple scales enable us to identify the small lesions while reducing many false alarms, and the multi-supervised scheme allows a better management of the unbalanced data. Compared to a single FCN (Dice score ~0.31), the performance on the testing dataset of our proposed networks achieved a Dice score of 0.78.
摘要
白质（WM）病变的识别和分割已被证明在诊断、治疗及神经学预后方面具有临床重要性。卷积神经网络（CNN）已在大负荷病变分割中取得成功，但对深部小WM病变和皮质下病变分割不敏感。我们提议使用多尺度和有监督的全卷积网络（FCN）对22例贫血患者的小WM病变进行分割。多尺度使我们能够识别小病变，同时减少许多误报，多监督方案则能更好地处理不平衡数据。与单个FCN（骰子系数约为0.31）相比，我们提出的网络在测试数据集上的性能达到了0.78的骰子系数。
相似文献
1
ORCHESTRAL FULLY CONVOLUTIONAL NETWORKS FOR SMALL LESION SEGMENTATION IN BRAIN MRI.用于脑磁共振成像中小病变分割的全卷积神经网络
Proc IEEE Int Symp Biomed Imaging. 2018 Apr;2018:889-892. doi: 10.1109/ISBI.2018.8363714. Epub 2018 May 24.
2
Segmentation of white matter hyperintensities using convolutional neural networks with global spatial information in routine clinical brain MRI with none or mild vascular pathology.利用卷积神经网络和全局空间信息对常规临床脑部 MRI（无或轻度血管病变）中的脑白质高信号进行分割。
Comput Med Imaging Graph. 2018 Jun;66:28-43. doi: 10.1016/j.compmedimag.2018.02.002. Epub 2018 Feb 17.
3
Ischemic Lesion Segmentation using Ensemble of Multi-Scale Region Aligned CNN.基于多尺度区域对齐卷积神经网络集成的缺血性病变分割。
Comput Methods Programs Biomed. 2021 Mar;200:105831. doi: 10.1016/j.cmpb.2020.105831. Epub 2020 Nov 12.
4
Deeply supervised 3D fully convolutional networks with group dilated convolution for automatic MRI prostate segmentation.基于深度监督的三维全卷积网络与分组空洞卷积在自动 MRI 前列腺分割中的应用。
Med Phys. 2019 Apr;46(4):1707-1718. doi: 10.1002/mp.13416. Epub 2019 Feb 19.
5
A Stacked Generalization of 3D Orthogonal Deep Learning Convolutional Neural Networks for Improved Detection of White Matter Hyperintensities in 3D FLAIR Images.用于改进 3D FLAIR 图像中脑白质高信号检测的三维正交深度学习卷积神经网络的堆叠泛化。
AJNR Am J Neuroradiol. 2021 Apr;42(4):639-647. doi: 10.3174/ajnr.A6970. Epub 2021 Feb 11.
6
Non-invasive multi-channel deep learning convolutional neural networks for localization and classification of common hepatic lesions.用于常见肝脏病变定位和分类的非侵入性多通道深度学习卷积神经网络
Pol J Radiol. 2021 Jul 20;86:e440-e448. doi: 10.5114/pjr.2021.108257. eCollection 2021.
7
Fully convolutional multi-scale residual DenseNets for cardiac segmentation and automated cardiac diagnosis using ensemble of classifiers.基于多尺度残差密集网络的全卷积神经网络模型及其在分类器集成中的应用，实现心脏分割和心脏疾病的自动化诊断。
Med Image Anal. 2019 Jan;51:21-45. doi: 10.1016/j.media.2018.10.004. Epub 2018 Oct 19.
8
Automatic segmentation of prostate cancer metastases in PSMA PET/CT images using deep neural networks with weighted batch-wise dice loss.使用带有加权批次骰子损失的深度神经网络对PSMA PET/CT图像中的前列腺癌转移灶进行自动分割。
Comput Biol Med. 2023 May;158:106882. doi: 10.1016/j.compbiomed.2023.106882. Epub 2023 Apr 4.
9
Skin lesion segmentation in dermoscopy images via deep full resolution convolutional networks.基于深度全分辨率卷积网络的皮肤镜图像皮损分割。
Comput Methods Programs Biomed. 2018 Aug;162:221-231. doi: 10.1016/j.cmpb.2018.05.027. Epub 2018 May 19.
10
Location Sensitive Deep Convolutional Neural Networks for Segmentation of White Matter Hyperintensities.位置敏感深度卷积神经网络在脑白质高信号分割中的应用
Sci Rep. 2017 Jul 11;7(1):5110. doi: 10.1038/s41598-017-05300-5.
引用本文的文献
1
A deep learning algorithm for white matter hyperintensity lesion detection and segmentation.一种用于检测和分割脑白质高信号病灶的深度学习算法。
Neuroradiology. 2022 Apr;64(4):727-734. doi: 10.1007/s00234-021-02820-w. Epub 2021 Oct 2.
2
Evaluating deep learning methods in detecting and segmenting different sizes of brain metastases on 3D post-contrast T1-weighted images.在三维增强后T1加权图像上评估深度学习方法用于检测和分割不同大小的脑转移瘤。
J Med Imaging (Bellingham). 2021 May;8(3):037001. doi: 10.1117/1.JMI.8.3.037001. Epub 2021 May 22.
3
Tract-specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke.镰状细胞病患者无显性卒中史的特定部位分析与神经认知功能。
Brain Behav. 2021 Mar;11(3):e01978. doi: 10.1002/brb3.1978. Epub 2021 Jan 12.
4
White matter has impaired resting oxygen delivery in sickle cell patients.白质在镰状细胞病患者中存在静息氧输送受损。
Am J Hematol. 2019 Apr;94(4):467-474. doi: 10.1002/ajh.25423. Epub 2019 Feb 21.
本文引用的文献
1
MSFCN-multiple supervised fully convolutional networks for the osteosarcoma segmentation of CT images.MSFCN——用于CT图像骨肉瘤分割的多监督全卷积网络
Comput Methods Programs Biomed. 2017 May;143:67-74. doi: 10.1016/j.cmpb.2017.02.013. Epub 2017 Feb 20.
2
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.
3
Fully Convolutional Networks for Semantic Segmentation.全卷积网络用于语义分割。
IEEE Trans Pattern Anal Mach Intell. 2017 Apr;39(4):640-651. doi: 10.1109/TPAMI.2016.2572683. Epub 2016 May 24.
4
Automatic Segmentation of MR Brain Images With a Convolutional Neural Network.基于卷积神经网络的磁共振脑图像自动分割。
IEEE Trans Med Imaging. 2016 May;35(5):1252-1261. doi: 10.1109/TMI.2016.2548501. Epub 2016 Mar 30.
5
Deep MRI brain extraction: A 3D convolutional neural network for skull stripping.深度磁共振成像脑部提取：一种用于去除颅骨的3D卷积神经网络。
Neuroimage. 2016 Apr 1;129:460-469. doi: 10.1016/j.neuroimage.2016.01.024. Epub 2016 Jan 22.
6
Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool.用于评估3D医学图像分割的指标：分析、选择与工具
BMC Med Imaging. 2015 Aug 12;15:29. doi: 10.1186/s12880-015-0068-x.
7
Automatic Detection of White Matter Hyperintensities in Healthy Aging and Pathology Using Magnetic Resonance Imaging: A Review.使用磁共振成像自动检测健康衰老和病理状态下的脑白质高信号：综述
Neuroinformatics. 2015 Jul;13(3):261-76. doi: 10.1007/s12021-015-9260-y.
8
Deep learning in neural networks: an overview.神经网络中的深度学习：综述。
Neural Netw. 2015 Jan;61:85-117. doi: 10.1016/j.neunet.2014.09.003. Epub 2014 Oct 13.
9
Controlled trial of transfusions for silent cerebral infarcts in sickle cell anemia.镰状细胞贫血症无症状脑梗死输血对照试验。
N Engl J Med. 2014 Aug 21;371(8):699-710. doi: 10.1056/NEJMoa1401731.
10
Adolescents with tetralogy of Fallot: neuropsychological assessment and structural brain imaging.法洛四联症青少年：神经心理学评估与脑结构成像
Cardiol Young. 2015 Feb;25(2):338-47. doi: 10.1017/S1047951114000031. Epub 2014 Feb 11.
Zotero 插件