Accelerating Cardiac Diffusion Tensor Imaging With a U-Net Based Model: Toward Single Breath-Hold

Outlier detection in cardiac diffusion tensor imaging: Shot rejection or robust fitting?

Cardiac diffusion tensor imaging (cDTI) is highly prone to image corruption, yet robust-fitting methods are rarely used. Single voxel outlier detection (SVOD) can overlook corruptions that are visually obvious, perhaps causing reluctance to replace whole-image shot-rejection (SR) despite its own deficiencies. SVOD's deficiencies may be relatively unimportant: corrupted signals that are not statistical outliers may not be detrimental. Multiple voxel outlier detection (MVOD), using a local myocardial neighbourhood, may overcome the shared deficiencies of SR and SVOD for cDTI while keeping the benefits of both. Here, robust fitting methods using M-estimators are derived for both non-linear least squares and weighted least squares fitting, and outlier detection is applied using (i) SVOD; and (ii) SVOD and MVOD. These methods, along with non-robust fitting with/without SR, are applied to cDTI datasets from healthy volunteers and hypertrophic cardiomyopathy patients. Robust fitting methods produce larger group differences with more statistical significance for MD, FA, and E2A, versus non-robust methods, with MVOD giving the largest group differences for MD and FA. Visual analysis demonstrates the superiority of robust-fitting methods over SR, especially when it is difficult to partition the images into good and bad sets. Synthetic experiments confirm that MVOD gives lower root-mean-square-error than SVOD.

Simultaneous non-contrast assessment of cardiac microstructure and perfusion in vivo in the human heart

BACKGROUND

Intravoxel incoherent motion (IVIM) imaging can provide information on cardiac microstructure and microvascular perfusion from a single examination. However, the spin echo-based approaches typically used for cardiac IVIM suffer from low sensitivity to changes in perfusion. The aim of this work was to develop a stimulated-echo (STEAM)-based method for IVIM and diffusion tensor cardiovascular magnetic resonance to simultaneously provide biomarkers of microstructure and perfusion in vivo in the human heart.

METHODS

Here we introduce a novel STEAM-IVIM sequence incorporating phase cycling to obtain true non-diffusion weighted images (b = 0 s/mm2). STEAM-IVIM imaging was performed at 20 b-values (0 to 1000 s/mm2) to enable accurate estimation of the IVIM parameters, and with six diffusion encoding directions to enable reconstruction of the diffusion tensor. 20 healthy subjects (8 female, median age 31 years) were imaged on a clinical 3T system with STEAM-IVIM. A simulation study was performed to investigate the optimal fitting algorithms for the IVIM parameters, which was subsequently used to create pixel-wise IVIM parameter maps for the in vivo acquisitions.

RESULTS

Good image quality across the myocardium was obtained for all b-values. Mean(±SD) IVIM parameter estimates were: diffusivity D = 0.83 ± 0.07 × 10-3 mm2/s, perfusion coefficient D* = 19.08 ± 6.48 × 10-3 mm2/s, perfusion fraction f = 19.72 ± 4.11%, and mean diffusion tensor parameters were: mean diffusivity = 0.88 ± 0.06 × 10-3 mm2/s, fractional anisotropy = 0.45 ± 0.04, absolute E2 angle = 55.29 ± 6.38º, helix angle gradient = -0.68 ± 0.18º/%.

CONCLUSION

Phase-cycled STEAM-IVIM enables fitting of cardiac diffusion tensor and perfusion parameters in healthy subjects and shows promise for the simultaneous detection of microstructural aberration and perfusion abnormalities in the presence of cardiac disease without the need for exogenous contrast agents.

Cardiac diffusion-weighted and tensor imaging: a Society for Cardiovascular Magnetic Resonance (SCMR) special interest group consensus statement

Thanks to recent developments in Cardiovascular magnetic resonance (CMR), cardiac diffusion-weighted magnetic resonance is fast emerging in a range of clinical applications. Cardiac diffusion-weighted imaging (cDWI) and diffusion tensor imaging (cDTI) now enable investigators and clinicians to assess and quantify the 3D microstructure of the heart. Free-contrast DWI is uniquely sensitized to the presence and displacement of water molecules within the myocardial tissue, including the intra-cellular, extra-cellular and intra-vascular spaces. CMR can determine changes in microstructure by quantifying: a) mean diffusivity (MD) -measuring the magnitude of diffusion; b) fractional anisotropy (FA) - specifying the directionality of diffusion; c) helix angle (HA) and transverse angle (TA) -indicating the orientation of the cardiomyocytes; d) E2A and E2A mobility - measuring the alignment and systolic-diastolic mobility of the sheetlets, respectively. This document provides recommendations for both clinical and research cDWI and cDTI, based on published evidence when available and expert consensus when not. It introduces the cardiac microstructure focusing on the cardiomyocytes and their role in cardiac physiology and pathophysiology. It highlights methods, observations and recommendations in terminology, acquisition schemes, post-processing pipelines, data analysis and interpretation of the different biomarkers. Despite the ongoing challenges discussed in the document and the need for ongoing technical improvements, it is clear that cDTI is indeed feasible, can be accurately and reproducibly performed and, most importantly, can provide unique insights into myocardial pathophysiology.

STEAM-SASHA: a novel approach for blood- and fat-suppressed native T1 measurement in the right ventricular myocardium

OBJECTIVE

The excellent blood and fat suppression of stimulated echo acquisition mode (STEAM) can be combined with saturation recovery single-shot acquisition (SASHA) in a novel STEAM-SASHA sequence for right ventricular (RV) native T1 mapping.

MATERIALS AND METHODS

STEAM-SASHA splits magnetization preparation over two cardiac cycles, nulling blood signal and allowing fat signal to decay. Breath-hold T1 mapping was performed in a T1 phantom and twice in 10 volunteers using STEAM-SASHA and a modified Look-Locker sequence at peak systole at 3T. T1 was measured in 3 RV regions, the septum and left ventricle (LV).

RESULTS

In phantoms, MOLLI under-estimated while STEAM-SASHA over-estimated T1, on average by 3.0% and 7.0% respectively, although at typical 3T myocardial T1 (T1 > 1200 ms) STEAM-SASHA was more accurate. In volunteers, T1 was higher using STEAM-SASHA than MOLLI in the LV and septum (p = 0.03, p = 0.006, respectively), but lower in RV regions (p > 0.05). Inter-study, inter-observer and intra-observer coefficients of variation in all regions were < 15%. Blood suppression was excellent with STEAM-SASHA and noise floor effects were minimal.

DISCUSSION

STEAM-SASHA provides accurate and reproducible T1 in the RV with excellent blood and fat suppression. STEAM-SASHA has potential to provide new insights into pathological changes in the RV in future studies.

Deep learning-based diffusion tensor cardiac magnetic resonance reconstruction: a comparison study

In vivo cardiac diffusion tensor imaging (cDTI) is a promising Magnetic Resonance Imaging (MRI) technique for evaluating the microstructure of myocardial tissue in living hearts, providing insights into cardiac function and enabling the development of innovative therapeutic strategies. However, the integration of cDTI into routine clinical practice poses challenging due to the technical obstacles involved in the acquisition, such as low signal-to-noise ratio and prolonged scanning times. In this study, we investigated and implemented three different types of deep learning-based MRI reconstruction models for cDTI reconstruction. We evaluated the performance of these models based on the reconstruction quality assessment, the diffusion tensor parameter assessment as well as the computational cost assessment. Our results indicate that the models discussed in this study can be applied for clinical use at an acceleration factor (AF) of × 2 and × 4 , with the D5C5 model showing superior fidelity for reconstruction and the SwinMR model providing higher perceptual scores. There is no statistical difference from the reference for all diffusion tensor parameters at AF × 2 or most DT parameters at AF × 4 , and the quality of most diffusion tensor parameter maps is visually acceptable. SwinMR is recommended as the optimal approach for reconstruction at AF × 2 and AF × 4 . However, we believe that the models discussed in this study are not yet ready for clinical use at a higher AF. At AF × 8 , the performance of all models discussed remains limited, with only half of the diffusion tensor parameters being recovered to a level with no statistical difference from the reference. Some diffusion tensor parameter maps even provide wrong and misleading information.

Mapping microarchitectural degeneration in the dilated ascending aorta with ex vivo diffusion tensor imaging

Aims

Thoracic aortic aneurysms (TAAs) carry a risk of catastrophic dissection. Current strategies to evaluate this risk entail measuring aortic diameter but do not image medial degeneration, the cause of TAAs. We sought to determine if the advanced magnetic resonance imaging (MRI) acquisition strategy, diffusion tensor imaging (DTI), could delineate medial degeneration in the ascending thoracic aorta.

Methods and results

Porcine ascending aortas were subjected to enzyme microinjection, which yielded local aortic medial degeneration. These lesions were detected by DTI, using a 9.4 T MRI scanner, based on tensor disorientation, disrupted diffusion tracts, and altered DTI metrics. High-resolution spatial analysis revealed that fractional anisotropy positively correlated, and mean and radial diffusivity inversely correlated, with smooth muscle cell (SMC) and elastin content (P < 0.001 for all). Ten operatively harvested human ascending aorta samples (mean subject age 61.6 ± 13.3 years, diameter range 29-64 mm) showed medial pathology that was more diffuse and more complex. Nonetheless, DTI metrics within an aorta spatially correlated with SMC, elastin, and, especially, glycosaminoglycan (GAG) content. Moreover, there were inter-individual differences in slice-averaged DTI metrics. Glycosaminoglycan accumulation and elastin degradation were captured by reduced fractional anisotropy (R2 = 0.47, P = 0.043; R2 = 0.76, P = 0.002), with GAG accumulation also captured by increased mean diffusivity (R2 = 0.46, P = 0.045) and increased radial diffusivity (R2 = 0.60, P = 0.015).

Conclusion

Ex vivo high-field DTI can detect ascending aorta medial degeneration and can differentiate TAAs in accordance with their histopathology, especially elastin and GAG changes. This non-destructive window into aortic medial microstructure raises prospects for probing the risks of TAAs beyond lumen dimensions.

Slice-specific tracking for free-breathing diffusion tensor cardiac MRI

BACKGROUND

Diffusion tensor cardiac magnetic resonance (DT-CMR) imaging has great potential to characterize myocardial microarchitecture. However, its accuracy is limited by respiratory and cardiac motion and long scan times. Here, we develop and evaluate a slice-specific tracking method to improve the efficiency and accuracy of DT-CMR acquisition during free breathing.

METHODS

Coronal images were obtained along with signals from a diaphragmatic navigator. Respiratory and slice displacements were obtained from the navigator signals and coronal images, respectively, and these displacements were fitted with a linear model to obtain the slice-specific tracking factors. This method was evaluated in DT-CMR examinations of 17 healthy subjects, and the results were compared with those obtained using a fixed tracking factor of 0.6. DT-CMR with breath-holding was used for reference. Quantitative and qualitative evaluation methods were used to analyze the performance of the slice-specific tracking method and the consistency between the obtained diffusion parameters.

RESULTS

In the study, the slice-specific tracking factors showed an upward trend from the basal to the apical slice. Residual in-plane movements were lower in slice-specific tracking than in fixed-factor tracking (RMSE: 2.748 ± 1.171 versus 5.983 ± 2.623, P < 0.001). The diffusion parameters obtained using slice-specific tracking were not significantly different from those obtained from breath-holding acquisition (P > 0.05).

CONCLUSION

In free-breathing DT-CMR imaging, the slice-specific tracking method reduced misalignment of the acquired slices. The diffusion parameters obtained using this approach were consistent with those obtained with the breath-holding technique.