Basic MRI physics for radiotherapy physicists

Abstract

CT has long been the modality of choice for radiotherapy treatment planning, because of its inherent lack of spatial distortion and its natural ability to provide electron-density information about tissues, directly feeding in to dose calculations. Nevertheless, MRI offers much better soft-tissue contrast than CT, with the added benefit of zero radiation dose. Advances in MRI technology have improved its geometric accuracy, while image segmentation allows dose-related parameters to be inferred, making MRI a viable option for radiotherapy planning. In MRI-guided radiotherapy, MR images are used to delineate tumour boundaries and guide the radiation distribution in real time. MRI uses signals generated by nuclear magnetic resonance (NMR) of magnetic hydrogen nuclei (protons). The frequency of the signal depends on the strength of the magnetic field, typically 1.5 tesla giving an NMR frequency of 64 MHz. Spatial information is obtained by magnetic field gradients, which alter the local magnetic field (and hence the resonant frequency) as a function of position; analysis of the frequency and phase of the measured NMR signals allows images to be produced in 2- or 3-dimensions. The geometric accuracy of the images is crucial to MR-based treatment planning, and depends on the homogeneity of the main magnetic field and the linearity and homogeneity of the magnetic field gradients. A plethora of MRI pulse sequences exist, which can be optimised to improve the results of image segmentation for dose calculations. This presentation will cover MRI concepts and how they impact on MR-based treatment-planning and on MR-guided therapy.