Student: Olivier Jaubert
1st supervisor: Claudia Prieto, King’s College London
2nd supervisor: Daniel Rueckert, Imperial College London
Cardiovascular disease (CVD) is the leading single cause of morbidity and mortality in the Western world, causing over 65.000 deaths every year in England. Magnetic Resonance Imaging (MRI) is a very promising non-invasive tool for early risk assessment, guidance of therapy and treatment monitoring of CVD. MRI has been shown to provide exquisite depiction of cardiac anatomy (spin density images) in clinical practice. More recently, quantitative mapping of magnetic relaxation properties (T1, T2 relaxation times) have been used for non-invasive tissue characterization [1]. T1 mapping techniques are enabling quantification of diffuse myocardial fibrosis while T2 mapping techniques allow the assessment of myocardial edema and inflammation.
The limitation of the current MRI scheme is that all these images (spin density, T1 map and T2 map) are acquired sequentially, thus resulting in long scan times. Magnetic Resonance Fingerprinting (MRF) is a novel, potentially revolutionary, technique that promises to overcome this problem [2]. This technique introduces new data acquisition, post processing and visualization approaches to achieve fully quantitative multiparametric MRI from a single acquisition. This is achieved by combining concepts of compressed measurements, dictionary learning and pattern recognition.
MRF has shown initial promising results in brain images [2] and breath hold abdominal images [3]. In this project we aim to develop M2D MRF for free-breathing multiparametric cardiac acquisitions. However several challenges need to be tackled to allow the application of MRF in cardiac imaging. Physiological cardiac and respiratory motion degrade the quality of the images and therefore need to be compensated or minimized during the acquisition. MRF has been shown to be robust to short non-periodic motion however further studies are required to observe the performance under complex cardiac and breathing motion. External and/or self-gated cardiac and respiratory signals will be incorporated in MRF to minimize remaining motion artefacts. Whole-heart acquisitions are highly desirable in cardiac imaging. So far MRF has only been shown for 2D images, thus new trajectories in combination with undersampled reconstruction will be required to extend this technique to multislice 2D (M2D).
This framework will be used to characterize the myocardium in healthy volunteers and patients with cardiovascular disease. The proposed approach will be validated by comparison with gold-standard spin density and parameter mapping images.
References:
[1] Salermo M, Kramer C. Advances in Parametric Mapping with CMR Imaging. JACC: Cardiovascular imaging 2013, 6: 806-822.
[2] Ma D, Gulani V, Seiberlich N et al. Magnetic Resonance Fingerprinting. Nature 2013, 495: 187-192.
[3] Chen Y, Jiang Y, Ma D et al. Magnetic Resonance Fingerprinting for Rapid Quantitative Abdominal Imaging. ISMRM 2014, 561.