Departments of Mechanical Engineering and Pharmacology, University of South Alabama, Mobile, AL, USA.
Department of Cardiology, College of Medicine, University of South Alabama, Mobile, AL, USA.
Br J Radiol. 2021 Apr 1;94(1120):20201101. doi: 10.1259/bjr.20201101. Epub 2021 Feb 24.
Left-ventricular (LV) strain measurements with the Displacement Encoding with Stimulated Echoes (DENSE) MRI sequence provide accurate estimates of cardiotoxicity damage related to chemotherapy for breast cancer. This study investigated an automated and supervised deep convolutional neural network (DCNN) model for LV chamber quantification before strain analysis in DENSE images.
The DeepLabV3 +DCNN with three versions of ResNet-50 backbone was designed to conduct chamber quantification on 42 female breast cancer data sets. The convolutional layers in the three ResNet-50 backbones were varied as non-atrous, atrous and modified, atrous with accuracy improvements like using Laplacian of Gaussian filters. Parameters such as LV end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were quantified, and myocardial strains analyzed with the Radial Point Interpolation Method (RPIM). Myocardial classification was validated with the performance metrics of accuracy, Dice, average perpendicular distance (APD) and others. Repeated measures ANOVA and intraclass correlation (ICC) with Cronbach's α (C-Alpha) tests were conducted between the three DCNNs and a vendor tool on chamber quantification and myocardial strain analysis.
Validation results in the same test-set for myocardial classification were accuracy = 97%, Dice = 0.92, APD = 1.2 mm with the modified ResNet-50, and accuracy = 95%, Dice = 0.90, APD = 1.7 mm with the atrous ResNet-50. The ICC results between the modified ResNet-50, atrous ResNet-50 and vendor-tool were C-Alpha = 0.97 for LVEF (55±7%, 54±7%, 54±7%, = 0.6), and C-Alpha = 0.87 for LVEDD (4.6 ± 0.3 cm, 4.6 ± 0.3 cm, 4.6 ± 0.4 cm, = 0.7).
Similar performance metrics and equivalent parameters obtained from comparisons between the atrous networks and vendor tool show that segmentation with the modified, atrous DCNN is applicable for automated LV chamber quantification and subsequent strain analysis in cardiotoxicity.
A novel deep-learning technique for segmenting DENSE images was developed and validated for LV chamber quantification and strain analysis in cardiotoxicity detection.
使用位移编码激励回波（DENSE）MRI 序列进行左心室（LV）应变测量，可以准确估计乳腺癌化疗相关的心脏毒性损伤。本研究旨在研究一种用于 DENSE 图像应变分析前的自动、监督的深度卷积神经网络（DCNN）模型，用于 LV 室量化。
设计了 DeepLabV3+DCNN 与三种版本的 ResNet-50 骨干网络，用于对 42 例女性乳腺癌数据集进行腔室量化。三种 ResNet-50 骨干网络中的卷积层分别为非扩张、扩张和改进型，改进型采用拉普拉斯高斯滤波器等方法提高准确性。量化参数包括 LV 舒张末期直径（LVEDD）和射血分数（LVEF），并采用径向点插值法（RPIM）分析心肌应变。采用准确性、Dice、平均垂直距离（APD）等性能指标对心肌分类进行验证。对三种 DCNN 和一种供应商工具在腔室量化和心肌应变分析方面进行了重复测量方差分析和组内相关系数（ICC）和 Cronbach's α（C-Alpha）检验。
在同一测试集中进行心肌分类的验证结果为：改进型 ResNet-50 的准确性=97%，Dice=0.92，APD=1.2mm；扩张型 ResNet-50 的准确性=95%，Dice=0.90，APD=1.7mm。改进型 ResNet-50、扩张型 ResNet-50 与供应商工具之间的 ICC 结果为：C-Alpha=0.97（55±7%、54±7%、54±7%、=0.6），LVEDD 的 C-Alpha=0.87（4.6±0.3cm、4.6±0.3cm、4.6±0.4cm、=0.7）。
与供应商工具相比，扩张网络之间的相似性能指标和等效参数表明，使用改良的、扩张的 DCNN 进行分割适用于自动化 LV 室量化和随后的心脏毒性应变分析。
开发并验证了一种用于 DENSE 图像分割的新型深度学习技术，用于 LV 室量化和心脏毒性检测中的应变分析。
