Automated segmentation of myocardial scar in late enhancement MRI using combined intensity and spatial information

Automatic classification of scar tissue in late gadolinium enhancement cardiac MRI for the assessment of left-atrial wall injury after radiofrequency ablation

Radiofrequency ablation is a promising procedure for treating atrial fibrillation (AF) that relies on accurate lesion delivery in the left atrial (LA) wall for success. Late Gadolinium Enhancement MRI (LGE MRI) at three months post-ablation has proven effective for noninvasive assessment of the location and extent of scar formation, which are important factors for predicting patient outcome and planning of redo ablation procedures. We have developed an algorithm for automatic classification in LGE MRI of scar tissue in the LA wall and have evaluated accuracy and consistency compared to manual scar classifications by expert observers. Our approach clusters voxels based on normalized intensity and was chosen through a systematic comparison of the performance of multivariate clustering on many combinations of image texture. Algorithm performance was determined by overlap with ground truth, using multiple overlap measures, and the accuracy of the estimation of the total amount of scar in the LA. Ground truth was determined using the STAPLE algorithm, which produces a probabilistic estimate of the true scar classification from multiple expert manual segmentations. Evaluation of the ground truth data set was based on both inter- and intra-observer agreement, with variation among expert classifiers indicating the difficulty of scar classification for a given a dataset. Our proposed automatic scar classification algorithm performs well for both scar localization and estimation of scar volume: for ground truth datasets considered easy, variability from the ground truth was low; for those considered difficult, variability from ground truth was on par with the variability across experts.

New automated Markov-Gibbs random field based framework for myocardial wall viability quantification on agent enhanced cardiac magnetic resonance images

A novel automated framework for detecting and quantifying viability from agent enhanced cardiac magnetic resonance images is proposed. The framework identifies the pathological tissues based on a joint Markov-Gibbs random field (MGRF) model that accounts for the 1st-order visual appearance of the myocardial wall (in terms of the pixel-wise intensities) and the 2nd-order spatial interactions between pixels. The pathological tissue is quantified based on two metrics: the percentage area in each segment with respect to the total area of the segment, and the trans-wall extent of the pathological tissue. This transmural extent is estimated using point-to-point correspondences based on a Laplace partial differential equation. Transmural extent was validated using a simulated phantom. We tested the proposed framework on 14 datasets (168 images) and validated against manual expert delineation of the pathological tissue by two observers. Mean Dice similarity coefficients (DSC) of 0.90 and 0.88 were obtained for the observers, approaching the ideal value, 1. The Bland-Altman statistic of infarct volumes estimated by manual versus the MGRF estimation revealed little bias difference, and most values fell within the 95% confidence interval, suggesting very good agreement. Using the DSC measure we documented statistically significant superior segmentation performance for our MGRF method versus established intensity-based methods (greater DSC, and smaller standard deviation). Our Laplace method showed good operating characteristics across the full range of extent of transmural infarct, outperforming conventional methods. Phantom validation and experiments on patient data confirmed the robustness and accuracy of the proposed framework.

Different algorithms for quantitative analysis of myocardial infarction with DE MRI: comparison with autopsy specimen measurements

RATIONALE AND OBJECTIVES

To compare two semiautomated methods for measurement of infarcted myocardium area on delayed contrast enhanced magnetic resonance imaging, with histopathology findings as standard of reference.

MATERIALS AND METHODS

Percentage area of myocardial infarction was measured in 10 Yorkshire landrace pigs manually and using two semiautomated methods. The first (standard deviation method) used two operator-selected regions of interest (ROIs) and nine different cutoff values (one to nine times the standard deviation of signal intensity in normal myocardium) to identify infarction. The second (threshold method) used threshold values based on percentages of maximum signal intensity to identify infarction. Results were compared with histopathology findings.

RESULTS

Difference between percentage area of infarction obtained with standard deviation method and autopsy specimens was in the range: -13.5% to +13.2%. With threshold method (thresholds from 30% to 90% of signal intensity), difference was -15% to +23%. Manual contouring underestimated infarcted area by 2% comparing to autopsy results. The best agreement between histopathology and semi-automated software was achieved for 4 standard deviations with standard deviation method: difference -0.45%, and for a percentage threshold of 70% (difference +0.67%) with threshold method. However, with standard deviation method, there was statistically significant difference between ROIs based on their location in viable myocardium: mean difference 1.7 ± 4%, P < .0001.

CONCLUSION

Semiautomated measurement of myocardial infarcted area on delayed enhanced magnetic resonance images performs well compared to autopsy. The threshold method, based on percentages of maximum signal intensity is preferable over standard deviation method, which is more susceptible to variability from location of ROIs within viable myocardium.

Toward magnetic resonance-guided electroanatomical voltage mapping for catheter ablation of scar-related ventricular tachycardia: a comparison of registration methods

INTRODUCTION

Integration of preprocedural delayed enhanced magnetic resonance imaging (DE-MRI) with electroanatomical voltage mapping (EAVM) may provide additional high-resolution substrate information for catheter ablation of scar-related ventricular tachycardias (VT). Accurate and fast image integration of DE-MRI with EAVM is desirable for MR-guided ablation.

METHODS AND RESULTS

Twenty-six VT patients with large transmural scar underwent catheter ablation and preprocedural DE-MRI. With different registration models and EAVM input, 3 image integration methods were evaluated and compared to the commercial registration module CartoMerge. The performance was evaluated both in terms of distance measure that describes surface matching, and correlation measure that describes actual scar correspondence. Compared to CartoMerge, the method that uses the translation-and-rotation model and high-density EAVM input resulted in a registration error of 4.32±0.69 mm as compared to 4.84 ± 1.07 (P <0.05); the method that uses the translation model and high-density EAVM input resulted in a registration error of 4.60 ± 0.65 mm (P = NS); and the method that uses the translation model and a single anatomical landmark input resulted in a registration error of 6.58 ± 1.63 mm (P < 0.05). No significant difference in scar correlation was observed between all 3 methods and CartoMerge (P = NS).

CONCLUSIONS

During VT ablation procedures, accurate integration of EAVM and DE-MRI can be achieved using a translation registration model and a single anatomical landmark. This model allows for image integration in minimal mapping time and is likely to reduce fluoroscopy time and increase procedure efficacy.