切换至 "中华医学电子期刊资源库"

中华医学超声杂志(电子版) ›› 2023, Vol. 20 ›› Issue (04) : 424 -429. doi: 10.3877/cma.j.issn.1672-6448.2023.04.008

心血管超声影像学

基于深度学习的超声心动图自动识别节段性室壁运动异常的研究
杨菲菲, 林锡祥, 陈亦新, 王秋霜, 张丽伟, 陈煦, 张梅青, 王淑华, 何昆仑()   
  1. 100048 北京,解放军总医院第六医学中心心内科;100039 北京,解放军总医院大数据中心
    100039 北京,解放军医学院
    101310 北京,北京安德医智科技有限公司
    100142 北京,解放军总医院第四医学中心健康医学科
    100048 北京,解放军总医院第六医学中心心内科
    100039 北京,解放军总医院大数据中心
  • 收稿日期:2021-11-02 出版日期:2023-04-01
  • 通信作者: 何昆仑
  • 基金资助:
    北京市自然科学基金(7202198); 国家自然科学基金(82202265)

Automatic detection of regional wall motion abnormalities by echocardiography based on deep learning

Feifei Yang, Xixiang Lin, Yixin Chen, Qiushuang Wang, Liwei Zhang, Xu Chen, Meiqing Zhang, Shuhua Wang, Kunlun He()   

  1. Department of Cardiology, The Sixth Medical Center of Chinese PLA General Hospital, 100048 Beijing, China; Medical Big Data Research Center, Chinese PLA General Hospital, 100039 Beijing, China
    Medical School of Chinese PLA, 100039 Beijing, China
    BioMind Technology, Zhongguancun Medical Engineering Center, 101310 Beijing, China
    Department of Cardiology, The Sixth Medical Center of Chinese PLA General Hospital, 100048 Beijing, China
    Medical Big Data Research Center, Chinese PLA General Hospital, 100039 Beijing, China
  • Received:2021-11-02 Published:2023-04-01
  • Corresponding author: Kunlun He
引用本文:

杨菲菲, 林锡祥, 陈亦新, 王秋霜, 张丽伟, 陈煦, 张梅青, 王淑华, 何昆仑. 基于深度学习的超声心动图自动识别节段性室壁运动异常的研究[J]. 中华医学超声杂志(电子版), 2023, 20(04): 424-429.

Feifei Yang, Xixiang Lin, Yixin Chen, Qiushuang Wang, Liwei Zhang, Xu Chen, Meiqing Zhang, Shuhua Wang, Kunlun He. Automatic detection of regional wall motion abnormalities by echocardiography based on deep learning[J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2023, 20(04): 424-429.

目的

探讨基于深度学习算法的超声心动图自动识别节段性室壁运动异常的效能。

方法

本研究回顾性收集了2015年6月至2019年9月在解放军总医院第四医学中心门诊及住院患者的超声心动图2274例作为训练集和验证集,其中包括心肌梗死患者1137例;另于2021年3月至2021年5月前瞻性收集1324例连续性超声心动图影像作为测试集,其中包括105例心肌梗死患者。本研究分为三个步骤,包括切面识别、左心室心肌分割以及室壁运动异常检测,并进一步比较了模型输入多个切面与输入单个切面对节段性室壁运动异常识别效能的差异。

结果

本研究神经卷积网络模型,对心尖四腔心切面(A4C),心尖两腔心切面(A2C)和心尖三腔心切面(A3C)的识别准确性分别为95%、98%、94%。心尖三个切面对左心室内膜分割的准确性均优于对心外膜的分割,且对心尖四腔心切面的分割准确性最佳(89.16%)。无论在内部验证集,还是外部测试集中,模型输入心尖三个切面对节段性室壁运动异常的识别效能均优于仅输入心尖四腔心单切面(ROC曲线下面积:0.942 vs 0.897;0.937 vs 0.828)。

结论

深度学习技术不仅可以自动识别超声心动图动态视频图像,并且可以识别节段性室壁运动异常,深度学习模型可以应用于临床实践,有助于提高超声的诊断效率。

Objectives

To investigate the efficiency of a deep learning (DL) framework to automatically analyze echocardiographic videos in detecting regional wall motion abnormalities.

Methods

A total of 2274 echocardiographic videos of outpatients and inpatients in the Fourth Medical Center of the PLA General Hospital from June 2015 to September 2019 were retrospectively collected as training and validation datasets, and 324 consecutive echocardiographic videos were prospectively collected as the test set, including 105 patients with myocardial infarction. We developed a three-stage DL framework, including 1) view classification, 2) left ventricular myocardial segmentation, and 3) regional wall motion abnormality detection. The difference in the recognition efficiency of segmental wall motion anomalies between the model based on videos in multiple views and those in a single view was then compared.

Results

The classification accuracy of the neural convolutional network model was 95%, 98%, and 94% for apical four chamber view (A4C), apical two chamber view (A2C), and apical three chamber view (A3C) videos, respectively. The accuracy of left ventricular endocardial segmentation based on videos in the three apical views was better than that of epicardial segmentation, and the segmentation result based on apical four chamber view videos was the best (89.16%). In both the internal validation dataset and external test dataset, the performance of the model based on videos in the three apical views was significantly better than that based on the single apical four chamber view videos (area under the curve: 0.942 vs 0.897; 0.937 vs 0.828).

Conclusions

The DL algorithm can not only automatically analyze echocardiographic videos, but also detect regional wall motion abnormalities. DL models can be applied in clinical practice to improve the diagnostic efficiency of echocardiography.

图1 节段性室壁运动识别模型结构示意图注:R2plus1D为残差网络结构:R=residual残差;D=dimension 维度;Feature Extrator为特征提取器;Adaptive avgPool为自适应池化;Classfication Module为分类模块;Normal为正常人;MI为心肌梗死患者
表1 训练集和测试集心肌梗死患者一般临床资料
图2 归一化混淆矩阵注:A4C-2D为心尖四腔心二维切面;A2C-2D为心尖二腔心二维切面;A3C-2D为心尖三腔心二维切面
表2 模型对于超声心动图切面分类以及左心室分割的效能分析
表3 模型输入心尖三个切面与输入心尖四腔心单切面识别节段性室壁运动异常的效能比较(内部验证集)
表4 模型输入心尖三个切面与输入心尖四腔心单切面识别节段性室壁运动异常的效能比较(外部测试集)
图3 心尖三个切面与心尖四腔心单切面识别室壁运动异常的ROC曲线(外部测试集)
1
Das N, Topalovic M, Janssens W. Artificial intelligence in diagnosis of obstructive lung disease: current status and future potential[J]. Curr Opin Pulm Med, 2018, 24(2): 117-123.
2
Rajkomar A, Dean J, Kohane I. Machine learning in medicine[J]. N Engl J Med, 2019, 380(26): 1347-1358.
3
Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: The Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC)[J]. Eur Heart J, 2018, 39(2):119-177.
4
Mitchell C, Rahko PS, Blauwet LA, et al. Guidelines for performing a comprehensive transthoracic echocardiographic examination in adults: recommendations from the American Society of Echocardiography[J]. J Am Soc Echocardiogr, 2019, 32(1):1-64.
5
Tran D, Wang H, Torresani L, et al. A closer look at spatiotemporal convolutions for action recognition[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2018.
6
Grewal PS, Oloumi F, Rubin U, et al. Deep learning in ophthalmology: a review[J]. Can J Ophthalmol, 2018, 53(4): 309-313.
7
Obermeyer Z, Emanuel EJ. Predicting the future-big data, machine learning, and clinical medicine[J]. N Engl J Med, 2016, 375: 1216-1219.
8
Alsharqi M, Woodward WJ, Mumith JA, et al. Artificial Intelligence and Echocardiography[J]. Echo Res Pract, 2018, 5(4): R115-R125.
9
Zhang J, Gajjala S, Agrawal P, et al. Fully automated echocardiogram interpretation in clinical practice[J]. Circulation, 2018, 138(16): 1623-1635.
10
Kusunose K, Abe T, Haga A, et al. A deep learning approach for assessment of regional wall motion abnormality from echocardiographic images [J]. JACC Cardiovasc Imaging, 2020, 13(2 Pt 1): 374-381.
[1] 何金梅, 尹立雪, 谭静, 张文军, 王锐, 任梅, 廖明娇. 超声心肌做功技术对2型糖尿病患者潜在左心室心肌收缩功能损伤的评价[J]. 中华医学超声杂志(电子版), 2023, 20(10): 1029-1035.
[2] 薛艳玲, 马小静, 谢姝瑞, 何俊, 夏娟, 何亚峰. 左心声学造影在急性心肌梗死合并室间隔穿孔中的应用价值[J]. 中华医学超声杂志(电子版), 2023, 20(10): 1036-1039.
[3] 武玺宁, 欧阳云淑, 张一休, 孟华, 徐钟慧, 张培培, 吕珂. 胎儿心脏超声检查在抗SSA/Ro-SSB/La抗体阳性妊娠管理中的应用[J]. 中华医学超声杂志(电子版), 2023, 20(10): 1056-1060.
[4] 杨水华, 何桂丹, 覃桂灿, 梁蒙凤, 罗艳合, 李雪芹, 唐娟松. 胎儿孤立性完全型肺静脉异位引流的超声心动图特征及高分辨率血流联合时间-空间相关成像的应用[J]. 中华医学超声杂志(电子版), 2023, 20(10): 1061-1067.
[5] 张宝富, 俞劲, 叶菁菁, 俞建根, 马晓辉, 刘喜旺. 先天性原发隔异位型肺静脉异位引流的超声心动图诊断[J]. 中华医学超声杂志(电子版), 2023, 20(10): 1074-1080.
[6] 张梅芳, 谭莹, 朱巧珍, 温昕, 袁鹰, 秦越, 郭洪波, 侯伶秀, 黄文兰, 彭桂艳, 李胜利. 早孕期胎儿头臀长正中矢状切面超声图像的人工智能质控研究[J]. 中华医学超声杂志(电子版), 2023, 20(09): 945-950.
[7] 赵红娟, 赵博文, 潘美, 纪园园, 彭晓慧, 陈冉. 应用多普勒超声定量分析正常中晚孕期胎儿左心室收缩舒张时间指数[J]. 中华医学超声杂志(电子版), 2023, 20(09): 951-958.
[8] 刘丹妮, 敖梦, 冉海涛, 李世玉, 秦芳. 三维超声心动图及二维斑点追踪成像对持续性心房颤动复律后双心房逆向重构的评估[J]. 中华医学超声杂志(电子版), 2023, 20(08): 827-835.
[9] 张璟璟, 赵博文, 潘美, 彭晓慧, 毛彦恺, 潘陈可, 朱玲艳, 朱琳琳, 蓝秋晔. 胎儿超声心动图测量McGoon指数在评价胎儿肺血管发育中的应用[J]. 中华医学超声杂志(电子版), 2023, 20(08): 860-865.
[10] 徐鹏, 李军, 高巍伦, 王峥, 庞珅, 李春妮, 朱霆. 快速旋转扫查法在胎儿超声心动图检查中的应用价值[J]. 中华医学超声杂志(电子版), 2023, 20(07): 761-766.
[11] 应康, 杨璨莹, 刘凤珍, 陈丽丽, 刘燕娜. 左心室心肌应变对无症状重度主动脉瓣狭窄患者的预后评估价值[J]. 中华医学超声杂志(电子版), 2023, 20(06): 581-587.
[12] 唐玮, 何融泉, 黄素宁. 深度学习在乳腺癌影像诊疗和预后预测中的应用[J]. 中华乳腺病杂志(电子版), 2023, 17(06): 323-328.
[13] 李晓阳, 刘柏隆, 周祥福. 大数据及人工智能对女性盆底功能障碍性疾病的诊断及风险预测[J]. 中华腔镜泌尿外科杂志(电子版), 2023, 17(06): 549-552.
[14] 邢晓伟, 刘雨辰, 赵冰, 王明刚. 基于术前腹部CT的卷积神经网络对腹壁切口疝术后复发预测价值[J]. 中华疝和腹壁外科杂志(电子版), 2023, 17(06): 677-681.
[15] 薛念余, 张盛敏, 吴凌恒, 沙蕾, 童揽月, 沈崔琴, 李朝军, 杜联芳. 研究血清胆红素对2型糖尿病患者心脏结构发生改变前心肌功能的影响[J]. 中华临床医师杂志(电子版), 2023, 17(9): 1004-1009.
阅读次数
全文


摘要