PhD Defense: Accelerated computational models for multi-parametric MRI reconstruction

This thesis focuses on advancing Magnetic Resonance Spin Tomography in Time-Domain (MR-STAT), a novel quantitative MRI (qMRI) technique. Comparing to conventional contrast-weighted MR techniques which produce contrast-weighted, qualitative MR images, MR-STAT is a novel quantitative MR technique, which uses a single short scan and produces the quantitative images which represent the specific tissue parameters, such as T1 and T2 relaxation times and proton density. These measurements provide a precise and objective assessment of tissue composition and pathology. This thesis focuses on two aspects of the technical development: the optimization of the reconstruction and acquisition processes.  

The reconstruction 

In this thesis, a new MR-STAT accelerated MR-STAT reconstruction algorithm is proposed. By leveraging artificial neural network and mathematical modelling strategies, the reconstruction time is significantly reduced to three minutes per slice. The algorithm has been applied in the first clinical demonstrator study, and the results for patients with various neurological diseases have been validated by radiologists.  

The acquisition 

Three-dimensional MR-STAT acquisition protocols have been developed in this thesis, enabling the acquisition of multi-parametric 3D quantitative volumes with lower noise and higher resolution. High-resolution 3D protocols have been tested on the brains, knees, and lower legs of healthy volunteers, with each scan taking between five (brain) and seven (extremities) minutes. These new protocols have greatly reduced scan times and improved image quality comparing to previous two-dimensional protocols, representing a major step towards the application of MR-STAT to various clinical applications in the future.