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 24 条
  • [1] Ventricle tracking in transesophageal echocardiography (TEE) images during cardiopulmonary resuscitation (CPR) using deep learning and monogenic filtering
    Ahn, Honggi
    Kim, Sun Ju
    Kang, Seungyoung
    Han, Junghun
    Hwang, Sung Oh
    Cha, Kyoung-Chul
    Yang, Sejung
    [J]. BIOMEDICAL ENGINEERING LETTERS, 2023, 13 (04) : 715 - 728
  • [2] Intracardiac vs Transesophageal Echocardiography for Left Atrial Appendage Occlusion With Watchman FLX in the US
    Ferro, Enrico G.
    Alkhouli, Mohamad
    Nair, Devi G.
    Kapadia, Samir R.
    Hsu, Jonathan C.
    Gibson, Douglas N.
    V. Freeman, James
    Price, Matthew J.
    Roy, Kristine
    Allocco, Dominic J.
    Yeh, Robert W.
    Piccini, Jonathan P.
    [J]. JACC-CLINICAL ELECTROPHYSIOLOGY, 2023, 9 (12) : 2587 - 2599
  • [3] Haukom T, 2019, IEEE INT ULTRA SYM, P1421, DOI [10.1109/ULTSYM.2019.8925996, 10.1109/ultsym.2019.8925996]
  • [4] Better With Ultrasound Transesophageal Echocardiography
    Jaidka, Atul
    Hobbs, Halley
    Koenig, Seth
    Millington, Scott J.
    Arntfield, Robert T.
    [J]. CHEST, 2019, 155 (01) : 194 - 201
  • [5] Left ventricle segmentation in transesophageal echocardiography images using a deep neural network
    Kang, Seungyoung
    Kim, Sun Ju
    Ahn, Hong Gi
    Cha, Kyoung-Chul
    Yang, Sejung
    [J]. PLOS ONE, 2023, 18 (01):
  • [6] LU-Net: A Multistage Attention Network to Improve the Robustness of Segmentation of Left Ventricular Structures in 2-D Echocardiography
    Leclerc, Sarah
    Smistad, Erik
    Ostvik, Andreas
    Cervenansky, Frederic
    Espinosa, Florian
    Espeland, Torvald
    Berg, Erik Andreas Rye
    Belhamissi, Mourad
    Israilov, Sardor
    Grenier, Thomas
    Lartizien, Carole
    Jodoin, Pierre-Marc
    Lovstakken, Lasse
    Bernard, Olivier
    [J]. IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL, 2020, 67 (12) : 2519 - 2530
  • [7] Deep Learning for Segmentation Using an Open Large-Scale Dataset in 2D Echocardiography
    Leclerc, Sarah
    Smistad, Erik
    Pedrosa, Joao
    Ostvik, Andreas
    Cervenansky, Frederic
    Espinosa, Florian
    Espeland, Torvald
    Berg, Erik Andreas Rye
    Jodoin, Pierre-Marc
    Grenier, Thomas
    Lartizien, Carole
    D'hooge, Jan
    Lovstakken, Lasse
    Bernard, Olivier
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2019, 38 (09) : 2198 - 2210
  • [8] Deep pyramid local attention neural network for cardiac structure segmentation in two-dimensional echocardiography
    Liu, Fei
    Wang, Kun
    Liu, Dan
    Yang, Xin
    Tian, Jie
    [J]. MEDICAL IMAGE ANALYSIS, 2021, 67
  • [9] Association between Transesophageal Echocardiography and Clinical Outcomes after Coronary Artery Bypass Graft Surgery
    MacKay, Emily J.
    Zhang, Bo
    Heng, Siyu
    Ye, Ting
    Neuman, Mark D.
    Augoustides, John G.
    Feinman, Jared W.
    Desai, Nimesh D.
    Groeneveld, Peter W.
    [J]. JOURNAL OF THE AMERICAN SOCIETY OF ECHOCARDIOGRAPHY, 2021, 34 (06) : 571 - 581
  • [10] A Shifting Paradigm The Role of Focused Cardiac Ultrasound in Bedside Patient Assessment
    Marbach, Jeffrey A.
    Almulfeh, Aws
    Di Santo, Pietro
    Simard, Trevor
    Jung, Richard
    Diemer, Gretchen
    West, Frances Mae
    Millington, Scott J.
    Mathew, Rebecca
    Le May, Michel R.
    Hibbert, Benjamin
    [J]. CHEST, 2020, 158 (05) : 2107 - 2118