Myocardial fibrosis is one of the most common manifestations of several cardiomyopathies and one of the main predictors of future cardiac events. Magnetic resonance imaging has been established as an emerging and powerful modality for the differentiation between healthy and non-healthy myocardium in a broad spectrum of cardiac diseases. Late gadolinium enhancement (LGE) is the reference technique for the visualization of myocardial scar and focal fibrosis, however it cannot visualize diffuse fibrosis. In contrast, quantitative myocardial T1 mapping allows the detection of both local and diffuse fibrosis, and it has been investigated as a potential new diagnostic tool for the assessment of different cardiomyopathies. Several T1 mapping techniques have been proposed in the last ten years, which are mostly restricted to two-dimensional (2D) imaging with limited spatial resolution, allowing the acquisition of a single slice in one breath-hold. In contrast, free-breathing three-dimensional (3D) quantitative T1 mapping benefits from a complete myocardial coverage with higher signal-to-noise ratio (SNR) and spatial resolution than a 2D acquisition. However, 3D acquisitions require longer scan time and consequently a free-breathing acquisition is mandatory. In this work a new 3D saturation recovery T1 mapping technique, called 3D SASHA, is presented, which provides whole-heart coverage during free-breathing. For respiratory motion compensation a 1D diaphragmatic navigator was used. However, this approach may only achieve a scan efficiency of approximately 50%, prolonging the scan time considerably. To shorten the acquisition a new solution is proposed, which instead combines the 3D SASHA imaging sequence with a 2D fat image navigator (fat-iNAV) for respiratory motion compensation. Motion estimation from the fat-iNAV may be more robust as image contrast using fat excitation is less affected by signal changes caused by the different saturation times compared to a water signal based navigator due to the short T1 of fat. The proposed fat-iNAV permits to estimate the respiratory motion of the heart with a higher accuracy, allowing 100% respiratory scan efficiency and predictable scan time. Another important aspect of the myocardial T1 mapping technique is to ensure an optimal trade-off between accuracy and precision. 3D SASHA imaging technique has shown higher accuracy in the estimation of the myocardial T1 than the clinically used inversion recovery based 2D MOLLI technique, but has lower precision. Here, a new approach is proposed, which combines the 3D SASHA sequence with a post-processing 3D Beltrami denoising method to improve the precision while maintaining the accuracy. After denoising, the precision of the 3D SASHA myocardial T1 mapping is substantially improved and is similar to that of 2D MOLLI T1 mapping, while preserving the higher accuracy and whole coverage of 3D SASHA. In addition, the proposed 3D Beltrami denoising approach is used to accelerate the 3D SASHA acquisition by reducing the number of T1-weighted images (from nine to three) used for the fitting process without affecting the accuracy and precision of the T1 map.
|Date of Award
|Rene Botnar (Supervisor) & Markus Henningsson (Supervisor)