文献速递心血管MR

优化深度学习用于心脏磁共振分割:自动层面范围分类的影响

Optimizing Deep Learning for Cardiac MRI Segmentation: The Impact of Automated Slice Range Classification.
PubMed 文献 · Acad Radiol · 2023-08-03

本研究在深度学习分割前加入自动层面范围分类步骤,基于公开多疾病、多视图、多中心右心室分割数据集(训练160例、验证40例、测试160例)评估3种分类模型和7种分割模型。联合分类(CBAM集成2D-CNN)与分割(带膨胀卷积块的2D-UNet)模型表现最优,左心室、右心室和心肌Dice分数分别为0.952、0.933和0.875,高于单独分割模型的0.949、0.925和0.867。联合模型与手动分割在双心室容积、射血分数和心肌质量上高度相关(0.92-0.99),临床参数平均绝对差(2.8-8.3 mL)在观察者间变异范围内,提示性能与手动标注相当。

要点

  • 联合自动层面范围分类与分割模型优于单独分割模型,左心室、右心室和心肌Dice分数分别为0.952、0.933和0.875。
  • 联合模型与手动分割在双心室容积、射血分数和心肌质量上高度相关(0.92-0.99)。
  • 自动与手动分割的临床参数平均绝对差为2.8-8.3 mL,在观察者间变异范围内。

核心结论

  • 在分割流程中整合初始自动层面范围分类步骤可改善基于深度学习的心脏腔室分割性能。

研究设计与样本

  • 采用公开的多疾病、多视图和多中心心脏磁共振右心室分割数据集,包含短轴电影数据,训练160例、验证40例、测试160例。
  • 研究了3种分类模型和7种分割模型,并对表现最佳的分割模型在有和无分类模型两种条件下进行评估。
  • 使用Dice分数、Hausdorff距离及临床指标(相关性评分和Bland-Altman图)比较自动与手动分割。

诊断标准与关键数值

  • 联合分类(CBAM集成2D-CNN)与分割(带膨胀卷积块的2D-UNet)模型Dice分数:左心室0.952、右心室0.933、心肌0.875。
  • 单独分割模型Dice分数:左心室0.949、右心室0.925、心肌0.867。
  • 联合模型与手动分割在双心室容积、射血分数和心肌质量上的相关性为0.92-0.99。
  • 自动与手动分割临床参数的平均绝对差为2.8-8.3 mL,在观察者间变异范围内。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

RATIONALE AND OBJECTIVES: 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.

MATERIALS AND METHODS: 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).

RESULTS: 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.

CONCLUSION: Integrating an initial automated slice range classification step into the segmentation process improves the performance of DL-based cardiac chamber segmentation.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Priya S, Dhruba DD, Perry SS, et al. Optimizing Deep Learning for Cardiac MRI Segmentation: The Impact of Automated Slice Range Classification. Acad Radiol. 2023. PMID: 37541826. doi:10.1016/j.acra.2023.07.008

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。