Related Experiment Video
Updated: Dec 26, 2025
10:53
Ultrasonic Assessment of Myocardial Microstructure
Published on: January 14, 2014
5.8K
Automatic quantification of left ventricular function by medical students using ultrasound
Jahn Frederik Grue1, Sigurd Storve2, Håvard Dalen2,3,4
1Department of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, NTNU, Prinsesse Kristinas gate 3, 7030, Trondheim, Norway. jfgrue@outlook.com.
BMC Medical Imaging
|March 18, 2020
Summary
Automatic analysis of echocardiograms can aid inexperienced users in quantifying left ventricular (LV) function. While automatic measurements of mitral annular plane systolic excursion (MAPSE) and peak velocities (S
Area of Science:
- Cardiology and Cardiovascular Imaging
- Medical Ultrasound Technology
- Artificial Intelligence in Medicine
Background:
- Echocardiography is crucial for assessing left ventricular (LV) function.
- Manual measurements of LV indices can be operator-dependent, especially for inexperienced users.
- Automatic analysis algorithms aim to standardize and simplify echocardiographic assessments.
Purpose of the Study:
- To develop and evaluate a fully automatic algorithm for measuring mitral annular plane systolic excursion (MAPSE) and peak systolic (S') and early diastolic (e') velocities.
- To assess the influence of user experience (medical students vs. clinicians) on the accuracy of automatic measurements.
- To determine the feasibility and agreement of automatic measurements with manual reference standards.
Main Methods:
- Seventy-five patients undergoing echocardiography were included.
- Manual reference measurements of MAPSE (M-mode) and S'/e' (color tissue Doppler imaging - cTDI) were obtained by clinicians.
- Medical students acquired cTDI recordings, which were then analyzed using a fully automatic algorithm for MAPSE, S', and e', and compared to manual references.
Main Results:
- Successful mitral annular tracking was achieved in 67% of student recordings and 84% of clinician recordings (p=0.007).
- Automatic measurements showed good agreement with manual references (Intraclass Correlation Coefficients: MAPSE 0.85, S' 0.89, e' 0.92) when tracking was correct.
- Agreement between automatic and manual measurements was similar for both students and clinicians, provided correct annular tracking.
Conclusions:
- The automatic algorithm provides good agreement with manual measurements for LV function indices when mitral annular tracking is successful.
- Image quality and adequate operator training are critical for the feasibility and accuracy of automatic echocardiographic analyses.
- Automatic measurements can assist less experienced users but do not eliminate the need for dedicated education and training in echocardiography.
