Fully Automated Convolutional Neural Network Method for Quantification of Breast MRI Fibroglandular Tissue and Background Parenchymal Enhancement
Fully Automated Convolutional Neural Network Method for Quantification of Breast MRI Fibroglandular Tissue and Background Parenchymal Enhancement
复制标题
DOI：
10.1007/s10278-018-0114-7
复制
发表时间：
2019-02-01
影响因子：
4.4
通讯作者：
Jambawalikar, Sachin
中科院分区：
文献类型：
--
作者：
Ha, Richard;Chang, Peter;Jambawalikar, Sachin
本研究的目的是开发一种全自动卷积神经网络（CNN）方法来量化乳腺MRI纤维腺组织（FGT）和背景实质增强（BPE）。一项机构审查委员会批准的回顾性研究评估了137例患者的1114个乳房体积，使用T1前对比、T1后对比和T1减影图像。首先，使用我们之前发表的量化方法，我们手动分割和计算FGT和BPE的数量，以建立地面真值参数。然后，在标准2D U-Net架构的基础上，开发并实现了一种新的3D CNN，用于体素预测全乳房和FGT边缘。在网络的折叠臂中，使用一系列尺寸为3 × 3 × 3的3D卷积滤波器进行标准的CNN分层特征提取。为了降低特征映射的维数，在所有方向上使用3 × 3 × 3卷积滤波器，步幅为2；总共使用了4个这样的操作。在网络的扩展臂中，使用一系列大小为3 × 3 × 3的卷积转置滤波器对每个中间层进行上采样。为了在多个分辨率下综合特征，在网络的伸缩臂和收缩臂之间引入了连接。L2正则化是为了防止过拟合。病例分为训练集（80%）和测试集（20%）。进行五重交叉验证。软件代码是在Linux工作站使用NVIDIA GTX Titan X GPU使用TensorFlow模块用Python编写的。在测试集中，全自动CNN方法量化FGT数量的准确率为0.813（交叉验证Dice得分系数），Pearson相关系数为0.975。对于量化BPE的数量，CNN方法的准确率为0.829，Pearson相关系数为0.955。我们的CNN网络能够在平均0.42秒内量化每个MRI病例的FGT和BPE。全自动CNN方法可用于量化MRI FGT和BPE。更大的数据集可能会改进我们的模型。
The aim of this study is to develop a fully automated convolutional neural network (CNN) method for quantification of breast MRI fibroglandular tissue (FGT) and background parenchymal enhancement (BPE). An institutional review board-approved retrospective study evaluated 1114 breast volumes in 137 patients using T1 precontrast, T1 postcontrast, and T1 subtraction images. First, using our previously published method of quantification, we manually segmented and calculated the amount of FGT and BPE to establish ground truth parameters. Then, a novel 3D CNN modified from the standard 2D U-Net architecture was developed and implemented for voxel-wise prediction whole breast and FGT margins. In the collapsing arm of the network, a series of 3D convolutional filters of size 3 x 3 x 3 are applied for standard CNN hierarchical feature extraction. To reduce feature map dimensionality, a 3 x 3 x 3 convolutional filter with stride 2 in all directions is applied; a total of 4 such operations are used. In the expanding arm of the network, a series of convolutional transpose filters of size 3 x 3 x 3 are used to up-sample each intermediate layer. To synthesize features at multiple resolutions, connections are introduced between the collapsing and expanding arms of the network. L2 regularization was implemented to prevent over-fitting. Cases were separated into training (80%) and test sets (20%). Fivefold cross-validation was performed. Software code was written in Python using the TensorFlow module on a Linux workstation with NVIDIA GTX Titan X GPU. In the test set, the fully automated CNN method for quantifying the amount of FGT yielded accuracy of 0.813 (cross-validation Dice score coefficient) and Pearson correlation of 0.975. For quantifying the amount of BPE, the CNN method yielded accuracy of 0.829 and Pearson correlation of 0.955. Our CNN network was able to quantify FGT and BPE within an average of 0.42 s per MRI case. A fully automated CNN method can be utilized to quantify MRI FGT and BPE. Larger dataset will likely improve our model.