...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Multimedia Tools and Applications >Voxelwise detection of cerebral microbleed in CADASIL patients by leaky rectified linear unit and early stopping
【24h】

Voxelwise detection of cerebral microbleed in CADASIL patients by leaky rectified linear unit and early stopping

机译：泄漏校正的线性单元提前停止立体定向检测CADASIL患者脑微出血。

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

It is important to detect cerebral microbleed voxels from the brain image of cerebral autosomal-dominant arteriopathy with subcortical infarcts and Leukoencephalopathy (CADASIL) patients. Traditional manual method suffers from intra-observe and inter-observe variability. In this study, we used the susceptibility weighted imaging (SWI) to scan 10 CADASIL patients and 10 healthy controls. We used slicing neighborhood processing (SNP) to extract “input” and “target” dataset from the 20 brain volumetric images. Afterwards, the undersampling technique was employed to handle the class-imbalanced problem. The single-hidden layer feedforward neural-network with scaled conjugate gradient was used as the classifier. We compared three activation functions: logistic sigmoid (LOSI), rectified linear unit (ReLU), and leaky rectified linear unit (LReLU). Early stopping and K-fold cross validation (CV) was used to avoid overfitting and statistical analysis. In the experiment, we generated 68,847 CMB voxels, and 68,829 non-CMB voxels. We observed that LReLU achieved the best result with a sensitivity of 93.05%, a specificity of 93.06%, and an accuracy of 93.06%. We also observed the effect of early stopping and K-fold CV. We found the optimal number of hidden neuron was 10 by grid searching method. Besides, our method performs better than three state-of-the-art methods. The results show our method is promising. In addition, LReLU is a better activation function that may replace traditional logistic sigmoid function in other applications.
机译：重要的是要从大脑常染色体显着性动脉病伴皮质下梗死和白细胞性脑病（CADASIL）患者的大脑图像中检测脑微出血体素。传统的手动方法存在观察内和观察间变异性。在这项研究中，我们使用药敏加权成像（SWI）扫描了10例CADASIL患者和10例健康对照。我们使用切片邻域处理（SNP）从20个大脑体积图像中提取“输入”和“目标”数据集。之后，采用欠采样技术来处理类不平衡问题。具有比例共轭梯度的单隐藏层前馈神经网络被用作分类器。我们比较了三种激活函数：逻辑乙状结肠（LOSI），整流线性单位（ReLU）和泄漏整流线性单位（LReLU）。早期停止和K折交叉验证（CV）用于避免过度拟合和统计分析。在实验中，我们生成了68,847个CMB体素和68,829个非CMB体素。我们观察到，LReLU以93.05％的灵敏度，93.06％的特异性和93.06％的准确性获得了最佳结果。我们还观察到了早期停止和K倍CV的效果。通过网格搜索法发现隐神经元的最佳数目为10。此外，我们的方法比三种最先进的方法表现更好。结果表明我们的方法是有前途的。此外，LReLU是一种更好的激活功能，可以在其他应用程序中替代传统的逻辑乙状结肠功能。

著录项

  • 来源
    《Multimedia Tools and Applications》 |2018年第17期|21825-21845|共21页
  • 作者

    Yu-Dong Zhang; Xiao-Xia Hou; Yi Chen; Hong Chen; Ming Yang; Jiquan Yang; Shui-Hua Wang;

  • 作者单位

    School of Computer Science and Technology, Nanjing Normal University,Hunan Provincial Key Laboratory of Network Investigational Technology,Department of Neurology, First Affiliated Hospital of Nanjing Medical University;

    Department of Neurology, First Affiliated Hospital of Nanjing Medical University;

    School of Computer Science and Technology, Nanjing Normal University,Hunan Provincial Key Laboratory of Network Investigational Technology;

    Department of Neurology, First Affiliated Hospital of Nanjing Medical University;

    Department of Radiology, Nanjing Children’s Hospital, Nanjing Medical University;

    School of Computer Science and Technology, Nanjing Normal University;

    School of Computer Science and Technology, Nanjing Normal University,Department of Electrical Engineering, The City College of New York, CUNY;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    CADASIL; Cerebral microbleed; Magnetic resonance imaging; Susceptibility weighted imaging; Class-imbalanced problem; Logistic sigmoid; Leaky rectified linear unit;

    机译：CADASIL;脑微出血;磁共振成像;药敏加权成像;类不平衡问题;逻辑乙状结肠;漏泄线性整流单元;

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号