Deep Learning Based Automatic Left Ventricle Segmentation from the Transgastric Short-Axis View on Transesophageal Echocardiography: A Feasibility Study

被引:0
|
作者
Tian, Yuan [1 ]
Qin, Wenting [2 ]
Zhao, Zihang [2 ]
Wang, Chunrong [1 ]
Tian, Yajie [1 ]
Zhang, Yuelun [1 ]
He, Kai [1 ]
Zhang, Yuguan [1 ]
Shen, Le [1 ]
Zhou, Zhuhuang [2 ]
Yu, Chunhua [1 ]
机构
[1] Chinese Acad Med Sci & Peking Union Med Coll, Peking Union Med Coll Hosp, Dept Anesthesiol, Beijing 100730, Peoples R China
[2] Beijing Univ Technol, Coll Chem & Life Sci, Dept Biomed Engn, Beijing 100124, Peoples R China
关键词
transesophageal echocardiography; deep learning; left ventricle segmentation; transgastric short-axis view; convolutional neural network; ULTRASOUND; SURGERY;
D O I
10.3390/diagnostics14151655
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Segmenting the left ventricle from the transgastric short-axis views (TSVs) on transesophageal echocardiography (TEE) is the cornerstone for cardiovascular assessment during perioperative management. Even for seasoned professionals, the procedure remains time-consuming and experience-dependent. The current study aims to evaluate the feasibility of deep learning for automatic segmentation by assessing the validity of different U-Net algorithms. A large dataset containing 1388 TSV acquisitions was retrospectively collected from 451 patients (32% women, average age 53.42 years) who underwent perioperative TEE between July 2015 and October 2023. With image preprocessing and data augmentation, 3336 images were included in the training set, 138 images in the validation set, and 138 images in the test set. Four deep neural networks (U-Net, Attention U-Net, UNet++, and UNeXt) were employed for left ventricle segmentation and compared in terms of the Jaccard similarity coefficient (JSC) and Dice similarity coefficient (DSC) on the test set, as well as the number of network parameters, training time, and inference time. The Attention U-Net and U-Net++ models performed better in terms of JSC (the highest average JSC: 86.02%) and DSC (the highest average DSC: 92.00%), the UNeXt model had the smallest network parameters (1.47 million), and the U-Net model had the least training time (6428.65 s) and inference time for a single image (101.75 ms). The Attention U-Net model outperformed the other three models in challenging cases, including the impaired boundary of left ventricle and the artifact of the papillary muscle. This pioneering exploration demonstrated the feasibility of deep learning for the segmentation of the left ventricle from TSV on TEE, which will facilitate an accelerated and objective alternative of cardiovascular assessment for perioperative management.
引用
收藏
页数:15
相关论文
共 50 条
  • [31] Automatic reorientation by deep learning to generate short-axis SPECT myocardial perfusion images
    Zhu, Fubao
    Wang, Guojie
    Zhao, Chen
    Malhotra, Saurabh
    Zhao, Min
    He, Zhuo
    Shi, Jianzhou
    Jiang, Zhixin
    Zhou, Weihua
    JOURNAL OF NUCLEAR CARDIOLOGY, 2023, 30 (05) : 1825 - 1835
  • [32] FULLY AUTOMATIC SEGMENTATION OF SHORT-AXIS CARDIAC R USING MODIFIED DEEP LAYER AGGREGATION
    Li, Zhongyu
    Lou, Yixuan
    Yan, Zhennan
    Al'Aref, Subhi
    Min, James K.
    Axel, Leon
    Metaxas, Dimitris N.
    2019 IEEE 16TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI 2019), 2019, : 793 - 797
  • [33] Automatic reorientation by deep learning to generate short-axis SPECT myocardial perfusion images
    Fubao Zhu
    Guojie Wang
    Chen Zhao
    Saurabh Malhotra
    Min Zhao
    Zhuo He
    Jianzhou Shi
    Zhixin Jiang
    Weihua Zhou
    Journal of Nuclear Cardiology, 2023, 30 : 1825 - 1835
  • [34] Use of fuzzy techniques for the automatic identification of left ventricle contours from multi-slice short-axis MR imaging
    Lalande, A
    Legrand, L
    Walker, PM
    Brunotte, F
    COMPUTERS IN CARDIOLOGY 1997, VOL 24, 1997, 24 : 149 - 152
  • [35] Segmentation of the Left Ventricle Using Active Contour Method with Gradient Vector Flow Forces in Short-Axis MRI
    Pieciak, Tomasz
    INFORMATION TECHNOLOGIES IN BIOMEDICINE, ITIB 2012, 2012, 7339 : 24 - 35
  • [36] An Image-Based Comprehensive Approach for Automatic Segmentation of Left Ventricle from Cardiac Short Axis Cine MR Images
    Su Huang
    Jimin Liu
    Looi Chow Lee
    Sudhakar K Venkatesh
    Lynette Li San Teo
    Christopher Au
    Wieslaw L. Nowinski
    Journal of Digital Imaging, 2011, 24 : 598 - 608
  • [37] An Image-Based Comprehensive Approach for Automatic Segmentation of Left Ventricle from Cardiac Short Axis Cine MR Images
    Huang, Su
    Liu, Jimin
    Lee, Looi Chow
    Venkatesh, Sudhakar K.
    Teo, Lynette Li San
    Au, Christopher
    Nowinski, Wieslaw L.
    JOURNAL OF DIGITAL IMAGING, 2011, 24 (04) : 598 - 608
  • [38] Dynamic pixel-wise weighting-based fully convolutional neural networks for left ventricle segmentation in short-axis MRI
    Wang, Zhongrong
    Xie, Lipeng
    Qi, Jin
    MAGNETIC RESONANCE IMAGING, 2020, 66 : 131 - 140
  • [39] Feasibility Study of Applying the Modified Hahn Nomenclature for Tricuspid Valve Leaflet Classification in Transthoracic Echocardiography Using the Subxiphoid Short-Axis View
    Wang, Hua
    Liu, Xiatian
    Zheng, Zhelan
    Wang, Bei
    ECHOCARDIOGRAPHY-A JOURNAL OF CARDIOVASCULAR ULTRASOUND AND ALLIED TECHNIQUES, 2025, 42 (03):
  • [40] A deep learning-based approach for automatic segmentation and quantification of the left ventricle from cardiac cine MR images
    Abdeltawab H.
    Khalifa F.
    Taher F.
    Alghamdi N.S.
    Ghazal M.
    Beache G.
    Mohamed T.
    Keynton R.
    El-Baz A.
    Computerized Medical Imaging and Graphics, 2020, 81