Department of Radiology, University of Iowa Carver College of Medicine, Iowa City, Iowa (S.P.).
Department of Electrical and Computer Engineering, University of Iowa, Iowa City, Iowa (D.D.D., M.J.).
Acad Radiol. 2024 Feb;31(2):503-513. doi: 10.1016/j.acra.2023.07.008. Epub 2023 Aug 3.
Cardiac magnetic resonance imaging is crucial for diagnosing cardiovascular diseases, but lengthy postprocessing and manual segmentation can lead to observer bias. Deep learning (DL) has been proposed for automated cardiac segmentation; however, its effectiveness is limited by the slice range selection from base to apex.
In this study, we integrated an automated slice range classification step to identify basal to apical short-axis slices before DL-based segmentation. We employed publicly available Multi-Disease, Multi-View & Multi-Center Right Ventricular Segmentation in Cardiac MRI data set with short-axis cine data from 160 training, 40 validation, and 160 testing cases. Three classification and seven segmentation DL models were studied. The top-performing segmentation model was assessed with and without the classification model. Model validation to compare automated and manual segmentation was performed using Dice score and Hausdorff distance and clinical indices (correlation score and Bland-Altman plots).
The combined classification (CBAM-integrated 2D-CNN) and segmentation model (2D-UNet with dilated convolution block) demonstrated superior performance, achieving Dice scores of 0.952 for left ventricle (LV), 0.933 for right ventricle (RV), and 0.875 for myocardium, compared to the stand-alone segmentation model (0.949 for LV, 0.925 for RV, and 0.867 for myocardium). Combined classification and segmentation model showed high correlation (0.92-0.99) with manual segmentation for biventricular volumes, ejection fraction, and myocardial mass. The mean absolute difference (2.8-8.3 mL) for clinical parameters between automated and manual segmentation was within the interobserver variability range, indicating comparable performance to manual annotation.
Integrating an initial automated slice range classification step into the segmentation process improves the performance of DL-based cardiac chamber segmentation.
心脏磁共振成像对于诊断心血管疾病至关重要，但冗长的后处理和手动分割可能导致观察者偏差。深度学习（DL）已被提出用于自动心脏分割；然而，其有效性受到从基底到心尖的切片范围选择的限制。
在这项研究中，我们在基于 DL 的分割之前集成了一个自动切片范围分类步骤，以识别基底到心尖的短轴切片。我们使用了公开的多疾病、多视图和多中心右心室分割在心脏 MRI 数据集，该数据集具有来自 160 个训练、40 个验证和 160 个测试病例的短轴电影数据。研究了三种分类和七种分割的 DL 模型。评估了表现最好的分割模型是否结合了分类模型。使用 Dice 评分和 Hausdorff 距离以及临床指标（相关评分和 Bland-Altman 图）对自动和手动分割进行模型验证。
联合分类（CBAM 集成 2D-CNN）和分割模型（2D-UNet 与扩张卷积块）表现出优越的性能，左心室（LV）的 Dice 评分达到 0.952，右心室（RV）达到 0.933，心肌达到 0.875，与独立分割模型（LV 为 0.949，RV 为 0.925，心肌为 0.867）相比。联合分类和分割模型与手动分割的双心室容积、射血分数和心肌质量具有高度相关性（0.92-0.99）。临床参数的平均绝对差异（2.8-8.3 毫升）在自动和手动分割之间，处于观察者内变异性范围内，表明与手动注释具有相当的性能。
将初始自动切片范围分类步骤集成到分割过程中可以提高基于 DL 的心脏腔室分割的性能。
