基于深度学习的图像融合方法提高钆延迟增强图像中心肌瘢痕的定量评估。
Improved Quantification of Myocardium Scar in Late Gadolinium Enhancement Images: Deep Learning Based Image Fusion Approach.
机构信息
Department of Medicine (Cardiovascular Division), Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts, USA.
Hypertrophic Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Massachusetts, USA.
出版信息
J Magn Reson Imaging. 2021 Jul;54(1):303-312. doi: 10.1002/jmri.27555. Epub 2021 Feb 17.
BACKGROUND
Quantification of myocardium scarring in late gadolinium enhanced (LGE) cardiac magnetic resonance imaging can be challenging due to low scar-to-background contrast and low image quality. To resolve ambiguous LGE regions, experienced readers often use conventional cine sequences to accurately identify the myocardium borders.
PURPOSE
To develop a deep learning model for combining LGE and cine images to improve the robustness and accuracy of LGE scar quantification.
STUDY TYPE
Retrospective.
POPULATION
A total of 191 hypertrophic cardiomyopathy patients: 1) 162 patients from two sites randomly split into training (50%; 81 patients), validation (25%, 40 patients), and testing (25%; 41 patients); and 2) an external testing dataset (29 patients) from a third site.
FIELD STRENGTH/SEQUENCE: 1.5T, inversion-recovery segmented gradient-echo LGE and balanced steady-state free-precession cine sequences ASSESSMENT: Two convolutional neural networks (CNN) were trained for myocardium and scar segmentation, one with and one without LGE-Cine fusion. For CNN with fusion, the input was two aligned LGE and cine images at matched cardiac phase and anatomical location. For CNN without fusion, only LGE images were used as input. Manual segmentation of the datasets was used as reference standard.
STATISTICAL TESTS
Manual and CNN-based quantifications of LGE scar burden and of myocardial volume were assessed using Pearson linear correlation coefficients (r) and Bland-Altman analysis.
RESULTS
Both CNN models showed strong agreement with manual quantification of LGE scar burden and myocardium volume. CNN with LGE-Cine fusion was more robust than CNN without LGE-Cine fusion, allowing for successful segmentation of significantly more slices (603 [95%] vs. 562 (89%) of 635 slices; P < 0.001). Also, CNN with LGE-Cine fusion showed better agreement with manual quantification of LGE scar burden than CNN without LGE-Cine fusion (%Scar = 0.82 × %Scar , r = 0.84 vs. %Scar = 0.47 × %Scar , r = 0.81) and myocardium volume (Volume = 1.03 × Volume , r = 0.96 vs. Volume = 0.91 × Volume , r = 0.91).
DATA CONCLUSION
CNN based LGE-Cine fusion can improve the robustness and accuracy of automated scar quantification.
LEVEL OF EVIDENCE
3 TECHNICAL EFFICACY: 1.
背景
由于心肌瘢痕与背景之间的对比度低和图像质量差，在晚期钆增强（LGE）心脏磁共振成像中对心肌瘢痕进行量化可能具有挑战性。为了解决模糊的 LGE 区域问题，有经验的读者通常使用传统的电影序列来准确识别心肌边界。
目的
开发一种深度学习模型，用于结合 LGE 和电影图像，以提高 LGE 瘢痕量化的稳健性和准确性。
研究类型
回顾性。
人群
总共 191 例肥厚型心肌病患者：1）来自两个站点的 162 例患者随机分为训练（50%，81 例）、验证（25%，40 例）和测试（25%，41 例）；2）来自第三个站点的 29 例外部测试数据集。
磁场强度/序列：1.5T，反转恢复分段梯度回波 LGE 和平衡稳态自由进动电影序列
评估
为心肌和瘢痕分割训练了两个卷积神经网络（CNN），一个带有 LGE-Cine 融合，一个不带 LGE-Cine 融合。对于具有融合功能的 CNN，输入为两个在匹配的心脏相位和解剖位置的对齐的 LGE 和电影图像。对于没有融合的 CNN，仅使用 LGE 图像作为输入。使用手动分割数据集作为参考标准。
统计检验
使用 Pearson 线性相关系数（r）和 Bland-Altman 分析评估 CNN 与手动量化 LGE 瘢痕负担和心肌体积的相关性。
结果
两种 CNN 模型均与手动量化的 LGE 瘢痕负担和心肌体积具有很强的一致性。具有 LGE-Cine 融合的 CNN 比不具有 LGE-Cine 融合的 CNN 更稳健，能够成功分割更多的切片（603[95%]与 562[89%]，共 635 个切片；P<0.001）。此外，具有 LGE-Cine 融合的 CNN 与手动量化的 LGE 瘢痕负担的一致性优于不具有 LGE-Cine 融合的 CNN（%Scar=0.82×%Scar，r=0.84 与%Scar=0.47×%Scar，r=0.81）和心肌体积（Volume=1.03×Volume，r=0.96 与 Volume=0.91×Volume，r=0.91）。
数据结论
基于 CNN 的 LGE-Cine 融合可以提高自动瘢痕量化的稳健性和准确性。
证据水平
3 技术功效：1。
Zotero 插件