Semi-automated myocardial segmentation of bright blood multi-gradient echo images improves reproducibility of myocardial contours and T2* determination

Right ventricular strain and volume analyses through deep learning-based fully automatic segmentation based on radial long-axis reconstruction of short-axis cine magnetic resonance images

OBJECTIVE

We propose a deep learning-based fully automatic right ventricle (RV) segmentation technique that targets radially reconstructed long-axis (RLA) images of the center of the RV region in routine short axis (SA) cardiovascular magnetic resonance (CMR) images. Accordingly, the purpose of this study is to compare the accuracy of deep learning-based fully automatic segmentation of RLA images with the accuracy of conventional deep learning-based segmentation in SA orientation in terms of the measurements of RV strain parameters.

MATERIALS AND METHODS

We compared the accuracies of the above-mentioned methods in RV segmentations and in measuring RV strain parameters by Dice similarity coefficients (DSCs) and correlation coefficients.

RESULTS

DSC of RV segmentation of the RLA method exhibited a higher value than those of the conventional SA methods (0.84 vs. 0.61). Correlation coefficient with respect to manual RV strain measurements in the fully automatic RLA were superior to those in SA measurements (0.5-0.7 vs. 0.1-0.2).

DISCUSSION

Our proposed RLA realizes accurate fully automatic extraction of the entire RV region from an available CMR cine image without any additional imaging. Our findings overcome the complexity of image analysis in CMR without the limitations of the RV visualization in echocardiography.

T2* assessment of the three coronary artery territories of the left ventricular wall by different monoexponential truncation methods

Objectives

This study aimed at evaluating left ventricular myocardial pixel-wise T2* using two truncation methods for different iron deposition T2* ranges and comparison of segmental T2* in different coronary artery territories.

Material and methods

Bright blood multi-gradient echo data of 30 patients were quantified by pixel-wise monoexponential T2* fitting with its R2 and SNR truncation. T2* was analyzed at different iron classifications. At low iron classification, T2* values were also analyzed by coronary artery territories.

Results

The right coronary artery has a significantly higher T2* value than the other coronary artery territories. No significant difference was found in classifying severe iron by the two truncation methods in any myocardial region, whereas in moderate iron, it is only apparent at septal segments. The R2 truncation produces a significantly higher T2* value than the SNR method when low iron is indicated.

Conclusion

Clear T2* differentiation between the three coronary territories by the two truncation methods is demonstrated. The two truncation methods can be used interchangeably in classifying severe and moderate iron deposition at the recommended septal region. However, in patients with low iron indication, different results by the two truncation methods can mislead the investigation of early iron level progression.

Cardiac T2 * mapping: Techniques and clinical applications

Cardiac T₂* mapping is a noninvasive MRI method that is used to identify myocardial iron accumulation in several iron storage diseases such as hereditary hemochromatosis, sickle cell disease, and β‐thalassemia major. The method has improved over the years in terms of MR acquisition, focus on relative artifact‐free myocardium regions, and T₂* quantification. Several improvement factors involved include blood pool signal suppression, the reproducibility of T₂* measurement as affected by scanner hardware, and acquisition software. Regarding the T₂* quantification, improvement factors include the applied curve‐fitting method with or without truncation of the signals acquired at longer echo times and whether or not T₂* measurement focuses on multiple segmental regions or the midventricular septum only. Although already widely applied in clinical practice, data processing still differs between centers, contributing to measurement outcome variations. State of the art T₂* measurement involves pixelwise quantification providing better spatial iron loading information than region of interest‐based quantification. Improvements have been proposed, such as on MR acquisition for free‐breathing mapping, the generation of fast mapping, noise reduction, automatic myocardial contour delineation, and different T₂* quantification methods. This review deals with the pro and cons of different methods used to quantify T₂* and generate T₂* maps. The purpose is to recommend a combination of MR acquisition and T₂* mapping quantification techniques for reliable outcomes in measuring and follow‐up of myocardial iron overload. The clinical application of cardiac T₂* mapping for iron overload's early detection, monitoring, and treatment is addressed. The prospects of T₂* mapping combined with different MR acquisition methods, such as cardiac T₁ mapping, are also described.

Level of Evidence: 4

Technical Efficacy Stage: 5

J. Magn. Reson. Imaging 2019.