Related Experiment Video
Updated: Apr 26, 2026
11:13
Quantification of Mouse Heart Left Ventricular Function, Myocardial Strain, and Hemodynamic Forces by Cardiovascular Magnetic Resonance Imaging
Published on: May 24, 2021
8.4K
Fast automatic myocardial segmentation in 4D cine CMR datasets.
Sandro Queirós1, Daniel Barbosa2, Brecht Heyde3
1Lab on Cardiovascular Imaging and Dynamics, KU Leuven, Belgium; ICVS/3B's - PT Government Associate Laboratory, Braga/Guimarães, Portugal.
Medical Image Analysis
|July 22, 2014
Summary
A new framework automatically segments the left ventricle (LV) in 3D+time cardiac MRI scans. This method accurately tracks LV contours throughout the cardiac cycle, offering robust and efficient analysis for cardiovascular imaging.
Area of Science:
- Cardiovascular Imaging
- Medical Image Analysis
- Computational Anatomy
Background:
- Accurate segmentation of the left ventricle (LV) is crucial for diagnosing cardiac conditions.
- Existing methods for 3D+time cardiac magnetic resonance (CMR) segmentation often face challenges with accuracy and computational efficiency.
- Automating the delineation of endocardial and epicardial contours throughout the cardiac cycle remains a significant task.
Purpose of the Study:
- To introduce a novel automatic 3D+time framework for segmenting left ventricle (LV) contours in cardiac magnetic resonance (CMR) datasets.
- To adapt and extend the B-spline Explicit Active Surfaces (BEAS) framework for robust segmentation of CMR images.
- To develop an efficient and accurate method for tracking LV surfaces across the cardiac cycle.
Main Methods:
- A three-block framework involving 2D mid-ventricular initialization, 3D end-diastolic segmentation, and temporal tracking.
- Adaptation of the B-spline Explicit Active Surfaces (BEAS) framework with dedicated energy terms for CMR properties.
- Extension of BEAS to a cylindrical space for 3D MR data topology, coupled with a fast stack initialization.
- Application of an anatomically constrained optical flow method for temporal LV surface tracking.
Main Results:
- The proposed framework demonstrated robust, efficient, and competitive performance in segmenting LV contours.
- Validation on 45 CMR datasets from the 2009 MICCAI LV segmentation challenge confirmed the method's accuracy.
- The approach achieved a favorable balance between segmentation accuracy and computational load.
Conclusions:
- The novel automatic 3D+time LV segmentation framework provides a significant advancement in CMR image analysis.
- The adapted BEAS framework and optical flow tracking offer a reliable solution for delineating LV endo- and epicardial borders.
- This method holds promise for improving the efficiency and accuracy of quantitative cardiovascular assessments.
