1 Department of Clinical Medicine - The MR Research Centre, Department of Clinical Medicine, Health, Aarhus University2 Department of Clinical Medicine - MTA, Department of Clinical Medicine, Health, Aarhus University3 Department of Clinical Medicine - Diagnostisk Center, Universitetsklinik for Innovative Patientforløb, Silkeborg, Department of Clinical Medicine, Health, Aarhus University4 Department of Clinical Medicine - Diagnostisk Center, Universitetsklinik for Innovative Patientforløb, Silkeborg, Department of Clinical Medicine, Health, Aarhus University
Real-time acquisition, reconstruction and interactively changing the slice position using magnetic resonance imaging (MRI) have been possible for years. However, the current clinical use of interactive real-time MRI is limited due to an inherent low spatial and temporal resolution. This PhD project seeks to implement and assess existing reconstruction algorithms using multi-processors of modern graphics cards and many-core computer processors and to cover some of the potential clinical applications which might benefit from using an interactive real-time MRI system. First an off-line, but interactive, slice alignment tool was used to support the notion that 3D blood flow quantification in the heart possesses the ability to obtain curves and volumes which are not statistical different from standard 2D flow. Secondly, the feasibility of an interactive real-time MRI system was exploited with regard to optimal sampling strategy for detecting motion in four different anatomies on two different MRI scanner brands. A fully implemented interactive real-time MRI system was exploited in a group of healthy fetuses and proved its eligibility as an alternative diagnostic tool for fetal imaging. Finally, the system was used for 3D motion tracking of the liver, and its use was proposed for future integrations of MRI scanners and linear accelerators in the field of radiotherapy treatment.