Radiotherapy is a non-invasive therapeutic modality that uses highly targeted radiation to treat cancer. The success of the treatment is to a large extent related to the precision of the radiation delivery. The precision of the radiation delivery depends, among others, on the visualization of the cancer before and during the radiation delivery. Traditionally the cancer is visualized with X-ray type of imaging techniques such as computed tomography (CT), which does not always provide sufficient image quality to precisely localize the cancer. To improve the visualization, radiotherapy departments are now more frequently using magnetic resonance imaging (MRI) to localize the cancer before the treatment, on diagnostic MRI systems, and during radiation delivery on hybrid MR-linac systems. The MRI scans provided by these systems are typically derived from radiology and are not necessarily tailored towards radiotherapy. In this thesis I present works on the design and implementation of MR scans that are optimized for specific radiotherapy applications, such as imaging in the presence of system hardware imperfections, high quality MR imaging while the patient is breathing and real-time estimation of the motion of moving tumors during irradiation. These dedicated MRI scans could potentially reduce motion-related imprecision of the radiation delivery while simultaneously providing improved integration and therefore acceleration of the radiotherapy workflow. The further development and integration of these new MRI scans could be important for the long term clinical impact of the use of MRI in radiotherapy.

Download Thesis