Search

This paper investigates thermal ablation of cancer tissue as a controls problem. The objective is to deliver a lethal dose of thermal energy to the cancer tissue without damaging critical nearby healthy tissue. This paper proposes a model predictive controller (MPC) to regulate thermal dose. The model predictive controller guarantees the safety of the healthy tissue surrounding the tumor. The cost function rewards killing cancer cells. Simulations comparing the proposed controller with the bang-bang controller used in practice demonstrate that the MPC controller kills more of the cancer tissue and requires less time than the bang-bang controller, while damaging less healthy tissue.

Developing tracers for MPI and related applications requires a thorough understanding of underlying relaxation processes. In X-Space MPI reconstruction, relaxation has a profound impact on the image through a decrease in SNR and direction-dependent blurring. To probe the relaxation of potential MPI tracers, Goodwill et al., have developed an X-Space Relaxometer that measures the MPI point spread function and accurately predicts image quality.[1,2] The present work introduces a new relaxometer with improved eddy current shielding to achieve a significant gain in sensitivity. A simulation of the enhanced shielding has shown to attenuate external interference by more than 40dB in the detection bandwidth.

MPI resolution and SNR are driven by the magnetic response of the iron oxide nanoparticles. Prior work has shown that magnetic nanoparticles respond differently to a wide range of excitation frequencies and amplitudes. Most prior work on MPI spectrometers and relaxometers have tuned the transmit coil at a single frequency.1-3 In contrast, in this work, we have constructed an agile frequency and amplitude relaxometer capable of reaching excitation fields of 0–60 mTpp across a continuous frequency range of DC-150 kHz without requirement of capacitors for tuning to discrete frequencies.

Magnetic Particle Imaging (mpi) is a promising new modality that images only a magnetic tracer, commonly super-paramagnetic iron oxide (spio) nanoparticles [1, 2]. During data acquisition information about the dc component of the native image is lost, and must be recovered using some a priori knowledge such as image continuity and positivity [3,4]. One method of recovery uses convex optimization techniques to determine the dc offsets that match the a priori knowledge. This method produces excellent images but significant time and memory resources to perform.

Previous work has explored the effects of nanoparticle relaxation from the system matrix and x-space reconstruction points of view [1-8]. Here we demonstrate how relaxation can be used to “colorize” an image based on relaxation mechanisms using x-space reconstruction [9] and a pixel-wise linear model.

MPI shows great promise for safe angiography, blood pool, and perfusion imaging. MPI images have unprecedented contrast that enables unambiguous measurements of tracer quantity and location [1-4]. At present, pre-clinical MPI resolution does not compare with pre-clinical MRI and CT, which routinely achieve better than 100-micron resolution. MPI's low resolution is primarily due to limitations in MPI hardware and magnetic nanoparticles. To begin to close this resolution gap, we are developing higher-field MPI gradients [3] and specialized MPI magnetic nanoparticles [4]. Since MPI resolution improves with stronger selection field gradients, we recently completed construction of a high-field MPI imager with a 7 T/m × 3.5 T/m × 3.5 T/m magnetic field gradient. Further, magnetic nanoparticles developed specifically for MPI show dramatic improvements in resolution and SNR over commercially available nanoparticles such as Resovist (BayerSchering) [4]. In this work, we demonstrate that combining these two breakthroughs enable improvements in MPI angiography images that begin to close the gap between MPI and pre-clinical MRI and CT.

An MPI spectrometer [1-3] measures the harmonics generated when a nonlinear SPIO is excited by a sinusoidal field. The harmonic peak spectrum allows for comparing SPIOs spatial resolution, since the harmonics are related to the modulation transfer function of the MPI imaging PSF [4-5]. It is also powerful to directly measure the point-spread function (PSF) by using the x-space MPI image reconstruction method [5-6], which has been shown to be equivalent to System Matrix reconstruction [7]. Here we describe a new design for an x-space relaxometer to increase sensitivity.