Related Experiment Video
Updated: Jun 10, 2026

Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
In-line automated tracking for ventricular function with magnetic resonance imaging.
Bo Li1, Yingmin Liu, Christopher J Occleshaw
1Auckland MRI Research Group, University of Auckland, Auckland, New Zealand.
An efficient nonrigid registration algorithm enables automatic feature tracking in cardiac MRI, significantly improving left ventricular volume analysis precision. This automated method enhances clinical assessment of ventricular function.
Area of Science:
- Medical Imaging
- Cardiovascular Imaging
- Biomedical Engineering
Background:
- Accurate assessment of left ventricular function is crucial for diagnosing and managing cardiovascular diseases.
- Traditional methods for analyzing cardiac magnetic resonance imaging (cMRI) data can be time-consuming and operator-dependent.
- Nonrigid registration algorithms offer potential for automating feature tracking in dynamic imaging sequences.
Purpose of the Study:
- To implement and evaluate an efficient nonrigid registration algorithm for in-line automatic feature tracking in steady-state free precession (SSFP) cine cardiac MRI.
- To assess the impact of this automated tracking on the precision of four-dimensional (4D) left ventricular function analysis.
- To compare the accuracy of left ventricular volume estimations with and without the use of the in-line automatic tracking.
Main Methods:
- An efficient nonrigid registration algorithm was implemented on an image reconstruction computer.
- The algorithm enabled in-line automatic tracking of image features in SSFP cine cardiac MRI sequences.
- Four-dimensional left ventricular function analysis was performed on data from 30 patients, both with and without the automated tracking results.
Main Results:
- In-line automatic tracking required approximately 10 ± 2 seconds per slice on the specified hardware.
- Clinical estimates of left ventricular end-diastolic volume (EDV) precision improved from 9 ml to 6 ml (p < 0.05) with automatic tracking.
- Left ventricular end-systolic volume (ESV) precision improved from 10 ml to 5 ml (p < 0.05) with automatic tracking compared to manual analysis.
Conclusions:
- In-line automatic tracking of image features using an efficient nonrigid registration algorithm is feasible and rapid.
- This automated approach significantly enhances the precision of clinical estimates of left ventricular volumes.
- The method shows considerable promise for facilitating more efficient and accurate clinical analysis of ventricular function in cardiac MRI.
Related Concept Videos
Magnetic Resonance Imaging
Imaging Studies for Cardiovascular System IV: CMRI
Imaging Studies IV: Magnetic Resonance Imaging

