Home // GLOBAL HEALTH 2019, The Eighth International Conference on Global Health Challenges // View article

Automated Measurement of Echocardiographic Global Longitudinal Strain

Authors:
Hisham Safawi
Mohamad Hajj-Hassan
Houssein Hajj-Hassan
Hassan M. Khachfe

Keywords: Global longitudinal strain; Echocardiography; Machine learning

Abstract:
Echocardiographic determination of Global Longitudinal Strain (GLS) by manual tracing of endocardial borders is time consuming and operator dependent, whereas visual assessment is inherently subjective. In this paper, the development of a fully automated software using machine learning-enabled image analysis is presented. For a total of 30 patients, apical 4, 2 and 3-chamber views were collected from center that assessed GLS using manual tracing. Manual tracing was done by same user to calculate user inimitability. In addition, datasets were saved in a centralized database, and machine learning-enabled software (AutoStrain, TomTec-Arena 1.2, TomTec Imaging Systems, Unterschleissheim, Germany) was applied for fully automated GLS measurements. AutoStrain measurements were feasible in 95% of studies and the average analysis time was less than 3 sec/ patient. Interclass correlation coefficients and Bland-Altman analysis revealed good ratio’s compared to manual tracking and user to user ratios. Fully automated analysis of echocardiography images provides rapid and reproducible assessment of left ventricular GLS compared to manual tracing.

Pages: 36 to 38

Copyright: Copyright (c) IARIA, 2019

Publication date: September 22, 2019

Published in: conference

ISSN: 2308-4553

ISBN: 978-1-61208-742-9

Location: Porto, Portugal

Dates: from September 22, 2019 to September 26, 2019