White matter hyperintensities segmentation using an ensemble of neural networks.
White matter hyperintensities segmentation using an ensemble of neural networks.
复制标题
使用神经网络集成进行白质高信号分割
DOI：
10.1002/hbm.25695
复制
发表时间：
2022-02-15
影响因子：
4.8
通讯作者：
Li Z
中科院分区：
文献类型：
--
作者：
Li X;Zhao Y;Jiang J;Cheng J;Zhu W;Wu Z;Jing J;Zhang Z;Wen W;Sachdev PS;Wang Y;Liu T;Li Z
白色高信号（WMH）是脑小血管病（CSVD）最常见的神经影像学标记。WMH的体积和位置是重要的临床测量。我们提出了一个使用深度全卷积网络和集成模型的管道，结合U‐Net，SE‐Net和多尺度特征，自动分割WMH并估计它们的体积和位置。我们在两个数据集中评估了我们的方法：一个是由60名患者组成的临床常规数据集（选自中国国家卒中登记中心，CNSR），另一个是由60名患者组成的研究数据集（选自MICCAI WMH挑战，MWC）。我们的管道的性能与四种免费提供的方法进行了比较：LGA，LPA，UBO检测器和U-Net，在各种指标方面。此外，为了访问模型泛化能力，选择并测试了另一个包括40名患者的研究数据集（来自澳大利亚老年双胞胎研究和悉尼记忆和衰老研究，OSM）。该管道在研究数据集和临床常规数据集中均实现了最佳性能，DSC显著高于其他方法（p <0.001），分别达到0.833和0.783。模型泛化能力的结果表明，在研究数据集上训练的模型（DSC = 0.736）比在临床数据集上训练的模型（DSC = 0.622）表现更好。我们的方法优于WMH分割中广泛使用的管道。该系统可以生成全脑、脑叶和解剖学自动标记WMH的图像和文本输出。此外，我们的方法的软件和模型可在https://www.nitrc.org/projects/what_v1上公开获得。
White matter hyperintensities (WMHs) represent the most common neuroimaging marker of cerebral small vessel disease (CSVD). The volume and location of WMHs are important clinical measures. We present a pipeline using deep fully convolutional network and ensemble models, combining U‐Net, SE‐Net, and multi‐scale features, to automatically segment WMHs and estimate their volumes and locations. We evaluated our method in two datasets: a clinical routine dataset comprising 60 patients (selected from Chinese National Stroke Registry, CNSR) and a research dataset composed of 60 patients (selected from MICCAI WMH Challenge, MWC). The performance of our pipeline was compared with four freely available methods: LGA, LPA, UBO detector, and U‐Net, in terms of a variety of metrics. Additionally, to access the model generalization ability, another research dataset comprising 40 patients (from Older Australian Twins Study and Sydney Memory and Aging Study, OSM), was selected and tested. The pipeline achieved the best performance in both research dataset and the clinical routine dataset with DSC being significantly higher than other methods (p < .001), reaching .833 and .783, respectively. The results of model generalization ability showed that the model trained on the research dataset (DSC = 0.736) performed higher than that trained on the clinical dataset (DSC = 0.622). Our method outperformed widely used pipelines in WMHs segmentation. This system could generate both image and text outputs for whole brain, lobar and anatomical automatic labeling WMHs. Additionally, software and models of our method are made publicly available at https://www.nitrc.org/projects/what_v1.
登录
查看更多内容
DOI：
10.1136/bmj.c3666
发表时间：
2010-07-26
期刊：
BMJ (Clinical research ed.)
影响因子：
--
作者：
Debette S;Markus HS
通讯作者：
Markus HS
影响因子：
4.2
作者：
Guerrero, R.;Qin, C.;Rueckert, D.
通讯作者：
Rueckert, D.
影响因子：
5.7
作者：
Griffanti L;Zamboni G;Khan A;Li L;Bonifacio G;Sundaresan V;Schulz UG;Kuker W;Battaglini M;Rothwell PM;Jenkinson M
通讯作者：
Jenkinson M
影响因子：
4.2
作者：
Jain, Saurabh;Sima, Diana M.;Ribbens, Annemie;Cambron, Melissa;Maertens, Anke;Van Hecke, Wim;De Mey, Johan;Barkhof, Frederik;Steenwijk, Martijn D.;Daams, Marita;Maes, Frederik;Van Huffel, Sabine;Vrenken, Hugo;Smeets, Dirk
通讯作者：
Smeets, Dirk
影响因子：
5.7
作者：
Li, Hongwei;Jiang, Gongfa;Menze, Bjoern
通讯作者：
Menze, Bjoern