Related Experiment Video
Updated: May 17, 2026
08:13
In Vivo Quantitative Assessment of Myocardial Structure, Function, Perfusion and Viability Using Cardiac Micro-computed Tomography
Published on: February 16, 2016
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic
Karolien Jaspers1, Hendrik G Freling, Kees van Wijk
1Department of Radiology, University Medical Centre Groningen, University of Groningen, PO Box 30.001, 9700 RB Groningen, The Netherlands.
The International Journal of Cardiovascular Imaging
|October 12, 2012
Summary
A new semi-automatic algorithm for cardiac MRI segmentation improves accuracy by excluding papillary muscles and trabeculae. This method enhances intra- and inter-observer agreement for cardiac function measurements compared to manual tracing.
Area of Science:
- Cardiovascular Imaging
- Medical Image Analysis
- Biomedical Engineering
Background:
- Accurate cardiac magnetic resonance (MR) imaging measurements of blood pool volume and ejection fraction are crucial for diagnosing heart conditions.
- Current manual contour tracing methods for cardiac MR analysis often include papillary muscles and trabeculae within the blood pool, potentially affecting accuracy.
- Novel segmentation algorithms are needed to refine the precision of MR-derived cardiac function parameters.
Purpose of the Study:
- To validate a novel semi-automatic segmentation algorithm for MR-derived cardiac volume and function measurements.
- To compare the performance of the semi-automatic method against the standard manual contour tracing technique.
- To assess the impact of excluding papillary muscles and trabeculae from the blood pool on cardiac parameter measurements.
Main Methods:
- Acquisition of multiphase 2D steady-state free precession short axis MR images in 12 subjects and an anthropomorphic heart phantom.
- Application of both manual contour tracing and a novel semi-automatic segmentation algorithm, using an epicardial contour as input.
- Comparison of cardiac parameters (end-diastolic volume, end-systolic volume, ejection fraction, myocardial volume) between the two methods and with true values in the phantom.
Main Results:
- The semi-automatic method, excluding papillary muscles and trabeculae, resulted in approximately 20 ml lower end-diastolic volume and end-systolic volume, and a 16% higher ejection fraction compared to manual methods in patients.
- Correlation between manual and semi-automatic measurements was high for all parameters (EDV: R = 0.99, ESV: R = 0.96, EF: R = 0.80, MV: R = 0.99).
- The semi-automatic method demonstrated significantly improved intra- and inter-observer agreement compared to the manual method.
Conclusions:
- The novel semi-automatic segmentation algorithm accurately measures cardiac MR-derived volumes and function.
- Excluding endoluminal muscular structures like papillary muscles and trabeculae improves the reliability and reproducibility of cardiac function measurements.
- This algorithm offers a more robust alternative to conventional manual contour tracing for cardiac MRI analysis.
