Related Experiment Video
Updated: Aug 17, 2026
In Vivo Quantitative Assessment of Myocardial Structure, Function, Perfusion and Viability Using Cardiac Micro-computed Tomography
Published on: February 16, 2016
Segmentation of non-viable myocardium in delayed enhancement magnetic resonance images
Arunark Kolipaka1, George P Chatzimavroudis, Richard D White
1Section of Cardiovascular Imaging, Division of Radiology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195, USA.
Insights
Five of six algorithms for segmenting non-viable myocardium in delayed enhancement MRI showed good agreement with manual methods. Semi-automatic algorithms performed best, though manual correction was sometimes needed for clinical use.
Area of Science:
- Cardiovascular Imaging
- Medical Image Analysis
- Cardiac MRI
Background:
- Delayed enhancement magnetic resonance imaging (DE-MRI) is crucial for assessing myocardial viability.
- Accurate segmentation of non-viable myocardium is essential for quantitative scar assessment.
- Automated algorithms aim to improve efficiency and reproducibility in DE-MRI analysis.
Purpose of the Study:
- To evaluate the performance of six different algorithms for segmenting non-viable left ventricular (LV) myocardium in DE-MRI.
- To compare automated segmentation results against manual thresholding as a reference standard.
Main Methods:
- Twenty-three patients with chronic ischemic heart disease underwent DE-MRI.
- Six automated thresholding algorithms were applied to DE images, differing in their intensity-based criteria (blood pool, viable myocardium, or both).
- Agreement with manual thresholding was assessed using Bland-Altman analysis to quantify bias and limits of agreement (LoA).
Main Results:
- Five of the six algorithms demonstrated a mean bias within +/-3% compared to manual segmentation.
- Limits of agreement were generally wide across all algorithms, ranging from 12% to 36%.
- The semi-automatic algorithms, Mean + 2SD(Semi) and Mean + 3SD(Semi), showed the best overall agreement with manual thresholding.
Conclusions:
- Five of the six evaluated algorithms are potentially satisfactory for clinical DE-MRI implementation.
- Semi-automatic approaches offer the most promising results for automated scar quantification.
- Manual review and correction may still be necessary for certain cases to ensure accuracy.
Purpose:
To evaluate six algorithms for segmenting non-viable left ventricular (LV) myocardium in delayed enhancement (DE) magnetic resonance imaging (MRI).
Methods:
Twenty-three patients with known chronic ischemic heart disease underwent DE-MRI. DE images were first manually thresholded using an interactive region-filling tool to isolate non-viable myocardium. Then, six thresholding algorithms, based on the image intensity characteristics of either LV blood pool (BP), viable LV myocardium, or both, were applied to each image. For the Mean-2SD(BP) algorithm, thresholds were equal to the mean BP intensity minus twice its standard deviation. For the Mean + 2SD(Semi), Mean + 3SD(Semi), Mean + 2SD(Auto), and Mean + 3SD(Auto) algorithms, thresholds equaled the mean intensity of viable myocardium plus twice (or thrice, as denoted by the name) the standard deviation of intensity (subscripts denote how these values were determined: automatic or semi-automatic). For the Minimum Intensity algorithm, the threshold equaled the minimum intensity between the BP and LV myocardium mean intensities. Percent Scar was defined as the ratio of non-viable to total myocardial pixels in each image. Agreement between each algorithm and manual thresholding was assessed using Bland-Altman analysis.
Results:
Mean Percent Scar was 25 +/- 16% by manual thresholding. Five of the six algorithms demonstrated mean bias within +/-3% (all except Mean+2SD(Auto)); however, limits of agreement (LoA) were large in general (range 12-36%). The best overall agreement was demonstrated by the Mean + 2SD(Semi) (bias, 0%; LoA, 12%) and Mean + 3SD(Semi)(bias, -3%; LoA, 14%) algorithms.
Conclusion:
On average, five of the six algorithms proved satisfactory for clinical implementation; however, in some images, manual correction of automatic results was necessary.
