Related Experiment Videos

Clinical validation of an automated boundary tracking algorithm on cardiac MR images

L A Latson1, K A Powell, B Sturm

  • 1Case Western Reserve University, Cleveland, OH, USA.

Insights

This study developed an automated algorithm for tracking left ventricle (LV) borders in cine-MRI scans. The algorithm accurately calculates LV volumes in healthy individuals and patients with heart conditions.

Area of Science:

  • Cardiovascular Imaging
  • Medical Image Analysis
  • Computational Cardiology

Background:

  • Accurate assessment of left ventricular (LV) volumes is crucial for diagnosing and managing cardiac diseases.
  • Manual tracing of LV borders in cine-MRI is time-consuming and subject to inter-observer variability.

Purpose of the Study:

  • To develop and validate an automated algorithm for tracking LV borders in cine-MRI gradient-echo temporal datasets.
  • To assess the algorithm's performance across diverse patient populations.

Main Methods:

  • An automated algorithm was developed to track endocardial and epicardial borders throughout the cardiac cycle in short-axis cine-MRI slices.
  • The algorithm utilized peak gradients along chords perpendicular to a centerline for border detection.
  • Validation involved comparing algorithm-derived LV volumes at end-systole (ES) with manual tracings.

Main Results:

  • The automated algorithm demonstrated high correlation with manual tracings for both endocardial (r2 = 0.72 - 0.98) and epicardial (r2 = 0.96 - 0.99) LV volumes at ES.
  • Favorable agreement was observed across healthy volunteers and patients with dilated cardiomyopathy, left ventricular hypertrophy, or left ventricular aneurysm.

Conclusions:

  • The developed automated algorithm provides a reliable and efficient method for LV border tracking and volume calculation from cine-MRI data.
  • This automated approach has the potential to improve the accuracy and consistency of cardiac MRI analysis in clinical practice.

Related Concept Videos