Related Experiment Video
Updated: Jun 20, 2026
06:34
Evaluation of Left Ventricular Structure and Function using 3D Echocardiography
Published on: October 28, 2020
MR evaluation of left ventricular volumes and function: threshold-based 3D segmentation versus short-axis planimetry
Kai Nassenstein1, Armin de Greiff, Peter Hunold
1Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany. Kai.Nassenstein@uni-due.de
Investigative Radiology
|September 3, 2009
Summary
This study demonstrates that a 3D threshold-based segmentation algorithm provides fast and reproducible left ventricular (LV) volume assessment in cardiac MRI. While it yields lower volumes than semi-automatic methods due to excluding trabeculations, it is highly accurate.
Area of Science:
- Cardiovascular Imaging
- Medical Image Analysis
- Quantitative MRI
Background:
- Accurate assessment of left ventricular (LV) volumes is crucial for diagnosing and managing cardiovascular diseases.
- Traditional methods for LV volume calculation in cardiac MRI can be time-consuming and operator-dependent.
Purpose of the Study:
- To evaluate a novel three-dimensional (3D) threshold-based, region-growing segmentation algorithm for LV volume assessment.
- To compare the accuracy and reproducibility of this automated algorithm against manual tracing and semi-automatic methods.
Main Methods:
- Acquisition of 2D cine steady-state free precession short-axis cardiac MRI scans in 44 patients.
- Application of an in-house developed 3D threshold-based, interslice region-growing segmentation algorithm.
- Comparison of results with manual tracing and semi-automatic contour detection (Argus software).
Main Results:
- The 3D threshold-based segmentation was feasible in 41/44 patients and showed no significant difference compared to manual tracing for end-diastolic (EDV), end-systolic (ESV), and stroke volumes (SV).
- Systematically lower LV volumes and ejection fraction were observed compared to short-axis planimetry with semi-automatic contour detection.
- Excellent reproducibility was demonstrated with low absolute differences between consecutive analyses, and significantly shorter evaluation times (60s vs. 278-669s).
Conclusions:
- Automated 3D threshold-based segmentation offers a fast and reproducible method for assessing LV volumes in cardiac MRI.
- The algorithm's exclusion of papillary muscles and trabeculations results in systematically lower LV volumes compared to semi-automatic methods.
