Magnetic particle imaging (MPI) is a tracer-based imaging modality with a variety of promising bio-medical applications including cancer diagnosis, cell tracking, and vascular imaging. It can potentially provide fast and sensitive mapping of superparamagnetic nanoparticle (SPN) distributions, and is expected to yield sub-millimeter resolution. These distributions are inferred from SPN responses to applied static and time-varying magnetic fields. The non-linear response of SPN magnetization to the time-varying component of the field induces time-dependent signals at harmonics of the fundamental in a detection coil. Signal localization is obtained by applying a strongly inhomogeneous static field that is arranged such that its amplitude is zero in certain `Field Free Regions' (FFRs). SPNs located near a FFR contribute significantly to the detected signal. The remaining SPNs contribute much less to the detected signal because they are exposed to a strong static magnetic field and their magnetization response is saturated. A map depicting the distribution of SPNs is then generated by systematically displacing the FFR over a desired Field Of View (FOV). Conventional approaches to MPI use the same current-driven time-varying magnetic fields to manipulate the FFR and to excite the SPNs. The resulting size of the FOV and the temporal resolution are proportional to the FFR-manipulation field amplitude and frequency, respectively. This fact, combined with restrictions imposed by health-related risks associated with high amplitude rapidly-varying magnetic fields represents a significant challenge to the field. In this thesis, I describe and demonstrate an alternative approach to MPI in which particle excitation and FFR manipulation are decoupled from one another. The additional degree of freedom enabled by this decoupling suggests new strategies for studying and exploiting contrast mechanisms, optimizing image quality and resolution, and device-size scaling. The prototype instrument I describe uses rotating arrays of permanent magnets to scan a Field Free Point (FFP) through the FOV, and current-driven oscillating magnetic fields to elicit non-linear magnetization responses from SPNs. Narrow-band phase-sensitive detection of these responses at one or more harmonics of the excitation field provides a rich source of information from which images can be reconstructed. Images generated from data acquired using this instrument are presented, demonstrating native resolutions of order one millimetre if the magnitude of the detected signal is employed. The resolution of these images can be improved at the expense of contrast using high harmonic components of the detected electromotive force. I also introduce a new imaging protocol, which we refer to as `phase-weighting', which substantially improves the spatial resolution of MP images through the use of phase information that is ultimately associated with SPN relaxation. Phase-weighted MP images with resolutions of order a few hundred micrometers, inferred from the width of the underlying point spread function, are obtained and presented. This represents a substantial advance in the state-of-the art within the field of MPI.
Copyright is held by the author.
This thesis may be printed or downloaded for non-commercial research and scholarly purposes.
Supervisor or Senior Supervisor
Thesis advisor: Hayden, Mike
Member of collection