Resting state functional magnetic resonance imaging (rsfMRI) has been widely used for measuring functional connectivity between cortical regions. However, there have been minimal reports of rsfMRI in white matter, presumably because of the sparse vasculature in white matter relative to gray, and the consistent failure to observe significant hemodynamic responses from tasks within white matter. In this study, we aimed to investigate and assess the nature of temporal variations in rsfMRI signals from human and monkey brains in white matter. Previous studies have reported that the correlations of time course signals in a resting state between voxels are anisotropic in white matter. We therefore constructed functional correlation tensors (FCTs) that quantify the functional relationships between neighboring voxels and their anisotropy in normal brains at rest, and compared these to underlying structural features. Furthermore, we elucidated the underlying biophysical mechanisms that account for their origins by assessing whether MRI signal fluctuations in white and gray matter vary for different baseline levels of neural activity. We found FCTs were capable of visualizing long range white matter tracts as well as short range sub-cortical fibers imaged at rest, suggesting temporal resting state signals may reflect intrinsic synchronizations of neural activity in white matter. Moreover, our monkey studies revealed that fractional power of rsfMRI signals are modulated similarly in regions of SI cortex, gray and white matter as neural baseline activity is varied. Our results imply that neural activity is encoded in white matter, and that that BOLD signal fluctuations in white matter may be detected in a resting state.

Given the prevalence of malaria in underdeveloped countries, specialized solutions must be developed for its diagnosis. The most common diagnostic throughout these regions is a lateral flow assay. While generally useful, these have a higher limit of detection than standard malaria diagnostics in the developed world. To help bridge this gap we are developing a system to concentrate malaria biomarkers on a lateral flow strip in order to lower its effective limit of detection. By using functionalized ferromagnetic micro-particles in a patient sample we can concentrate biomarkers and then transfer them to a surface leaving behind the bulk fluid. One of the first steps towards this is to understand parameters affecting particle transfer from an idealized fluid and across an air gap to a surface below. We were able to transfer various bead across distances of over 3mm for samples over 300?g when holding the sample orthogonal to the surface and over 200?g when holding the sample at a 45° angle to the surface. Bubble exchange between the seems to be the limiting factor driving bead transfer and occurs more readily in 45° samples allowing bead transfer to occur in an unimpeded fashion leading to more easily discernable trends within these samples.

<p>Prostate and breast cancers are two of the most common types of cancer in the United States, and those cancers metastasize to bone in more than two thirds of patients. Recent evidence suggests that thermal therapy is effective at treating metastatic bone cancer. For example, thermal therapy enables targeted drug delivery to bone, ablation of cancer cells in bone marrow, and palliation of bone pain. Thermal therapy of bone metastases would be greatly improved if it were possible to image the temperature of the tissue surrounding the disease, which is usually red bone marrow (RBM). Unfortunately, current thermal imaging techniques are inaccurate in RBM.</p><p>This dissertation shows that many of the difficulties with thermal imaging of RBM can be overcome using a magnetic resonance phenomenon called an intermolecular multiple quantum coherence (iMQC). Herein, iMQCs are detected with a magnetic resonance imaging (MRI) pulse sequence called multi-spin-echo HOMOGENIZED with off resonance transfer (MSE-HOT). Compared to traditional methods, MSE-HOT provided ten-fold more accurate images of temperature change. Furthermore, MSE-HOT was translated to a human MRI scanner, which enabled imaging of RBM temperature during heating with a clinical focused ultrasound applicator. In summary, this dissertation develops a MRI technique that enables thermal imaging of RBM during thermal therapy of bone metastases.</p> / Dissertation

<p>In this work, I investigate the mechanical response of the liver to increasing pressure in the portal vein using ultrasonic approaches. In advancing liver disease, portal venous pressure increases lead to severe clinical problems and death. Monitoring these pressure increases can predict patient outcomes and guide treatment. Current methods for measurement of portal venous pressure are invasive, expensive, and therefore are rarely repeated. Ultrasonic methods show promise because they are noninvasive, but traditional ultrasound images and doppler measurements do not yield accurate repeatable measures of hepatic pressure. However, increases in portal venous pressure have been associated with higher estimates of liver stiffness using ultrasound-based shear wave speed estimation algorithms. These quantitative estimates of shear wave speed may provide a mechanism for noninvasive hepatic pressure characterization, but they cannot currently be distinguished from the increases in shear wave speed estimates that are also observed in patients with normal portal venous pressures with advancing liver diseases. Thus, a better understanding of the mechanisms by which hepatic pressure modulates estimates of liver stiffness could provide information needed to distinguish increasing hepatic pressure from advancing brosis stage. This work is devoted to identifying and characterizing the underlying mechanism behind the observed increases in hepatic shear wave speed with pressurization.</p><p>Two experiments were designed in order to dene the mechanical properties of liver tissue that underlie the observed increase in shear wave speeds with increasing portal venous pressure. First, the behavior of the liver was shown to be nonlinear (or strain-dependent) by comparing stiness estimates in livers that were free to expand and constrained from expansion at increasing hepatic pressures. Shear wave speeds were observed to increase only in the unconstrained case in which the liver was observed to qualitatively deform. Second, the deformation of the liver was quantied using a clinical scanner and 3-D transducer to generate estimates of axial strain during pressurization. Axial strain was found to increase with elevation in portal venous pressure. This axial expansion of the liver also corresponded to increases in shear wave speed estimates with portal venous pressure.</p><p>The techniques developed herein were used to elucidate mechanical properties of the pressurized liver by concurrent ultrasound-based quantication of hepatic deformation and stiffness. This work shows that increasing shear wave speed estimates with hepatic pressurization are associated with increases in hepatic axial strain measurements. These results provide the basis for quantifying the relationship between pressurization and hepatic strain, laying the foundation for hyperelastic material modeling of the liver. Such nonlinear mechanical models can provide the basis for noninvasive characterization of hepatic pressure using stiffness metrics in the future.</p> / Dissertation

Electrostatic complexation of a cationic MAPKAP kinase 2 inhibitory (MK2i) peptide with the anionic, pH-responsive polymer poly(propylacrylic acid) (PPAA) yields MK2i nano-polyplexes (MK2i-NPs) that significantly increase peptide uptake and intracellular retention. This study focused on elucidating the mechanism of MK2i-NP cellular uptake and intracellular trafficking in vascular smooth muscle cells. Small molecule inhibition of various endocytic pathways showed that MK2i-NP cellular uptake involves both macropinocytosis and clathrin mediated endocytosis, whereas the free peptide utilizes clathrin mediated endocytosis alone for cell entry. Scanning electron microscopy studies revealed that MK2i-NPs, but not free MK2i peptide, induce cellular membrane ruffling consistent with macropinocytosis. TEM confirmed that MK2i-NPs induce macropinosome formation and achieve MK2i endo-lysosomal escape and cytosolic delivery. Finally, a novel technique based on recruitment of Galectin-8-YFP was developed and utilized to demonstrate that MK2i-NPs cause endosomal disruption within 30 minutes of uptake. These new insights on the relationship between NP physicochemical properties and cellular uptake and trafficking can potentially be applied to further optimize the MK2i-NP system and more broadly toward the rational engineering of nano-scale constructs for the intracellular delivery of biologic drugs.

Photothermal OCT (PT-OCT) is an emerging molecular imaging technique that occupies a spatial imaging regime between microscopy and whole body imaging. PT-OCT has been demonstrated in vitro, ex vivo, and in vivo on multiple contrast agents such as gold nanoparticles and indocyanine green. PT-OCT would benefit from a theoretical model to optimize imaging parameters and test image processing algorithm. Past models have focused on individual components of the PT-OCT signal, but a comprehensive model still had to be assembled. We propose the first PT-OCT model to replicate an A-scan in homogeneous and layered samples. Our predictions were validated experimentally in silicone phantoms. We also propose the PT-CLEAN algorithm to reduce phase-accumulation and shadowing, two artefacts found in PT-OCT images and demonstrate this algorithm on phantom and in vivo images.

Simultaneous multislice (SMS) imaging is a scan acceleration method where mul- tiple slices are simultaneously excited using a multiband pulse and the aliased slice images are separated in reconstruction using the receive coils sensitivity maps. At high main field strengths, SMS brain imaging suffers from artifacts caused by non- uniform and subject-dependent transmit RF fields and large magnetic susceptibility differences near air-tissue interfaces such as the frontal sinus and the middle ear. Another significant engineering challenge is the increase in peak power of multiband pulses with the number of excited slices. In this research work, we propose novel radiofrequency pulses and pulse sequences to address these SMS imaging problems. Low peak power multiband spokes excitation pulses are proposed to mitigate the image shading artifacts caused by inhomogeneous transmit RF field in multiple si- multaneously excited slices. Results from simulations and in vivo experiments at 7 T demonstrate that images excited using multiband spokes pulses have reduced center brightening artifact than conventional multiband pulses. We propose a novel pulse sequence called multispectral z-shim to reduce the through-plane signal loss artifact in structural and functional MR imaging. In vivo experiments show that the multispectral z-shim sequence recovers signal in regions of susceptibility difference in multiple brain regions while maintaining signal elsewhere. To reduce the peak power of conventional pulses, we present a method to design root-flipped multiband pulses. Simulations and experiments demonstrate that for a fixed peak amplitude, the root-flipped pulses excite the desired slices with a pulse duration lower than that of pulses proposed earlier. The work presented in this dissertation will improve high field SMS imaging research in areas such as functional MRI, susceptibility-weighted imaging and diffusion-weighted imaging.

Sensitive and specific noninvasive in vivo imaging of contrast agents and endogenous molecules can supply molecular and functional information in animal models, providing essential insight into mechanisms of disease formation and progression, drug delivery, and treatment response. In cancer in particular, high resolution imaging is essential for capturing spatial heterogeneities in molecular expression and the tumor microenvironment that cause significant barriers to treatment efficacy and drug delivery. Optical coherence tomography (OCT) fills the niche of cellular-level resolution and penetration depths in tissue that exceed those obtained with microscopy, an attractive regime for imaging mouse models of cancer. In this dissertation, photothermal OCT (PTOCT), a functional extension of OCT, was developed for in vivo imaging of a variety of contrast agents and drug delivery vectors in live animals. The PTOCT signal was thoroughly characterized in phantoms and compared to theory, followed by a demonstration of picomolar sensitivity to gold nanorod contrast agents. Gold nanorods at physiologically relevant concentrations were then identified from within a live mouse at depths exceeding the standard limits of high resolution optical microscopy. Then, heterogeneities in gold nanorod delivery to tumors were imaged in the context of tissue and vessel morphology, demonstrating the utility of PTOCT as part of a powerful multimodality imaging platform for the development of nanomedicines and drug delivery technologies. The uptake of gold nanorods into mouse mammary tumors were tracked in three dimensions over 24 hours, and the specificity of the PTOCT signal was verified using multiphoton microscopy. Finally, photothermal optical lock-in optical coherence tomography (poli-OCT) was used to increase system throughput and allow for real time photothermal imaging. In vivo poli-OCT of indocyanine green identified lymphatic vessels in a mouse ear, and also identified picomolar concentrations of gold nanorods in subcutaneous injections at frame rates ten times faster than previously reported. Overall, the development of in vivo PTOCT combined with existing morphological and hemodynamic imaging capabilities of OCT will enable more comprehensive studies of drug delivery and molecular expression in mouse models of disease, particularly cancer.

The objective of this dissertation was to develop an experimental approach and supporting software for performing and interpreting the results of micoarray-based experiments, as well as apply this approach to an experimental model of angiogenesis and blood vessel development. When this project was initiated microarray technology was in its infancy and the standard experimental design was to hybridize two samples against each other and report intensity ratios that were greater than two-fold. In order to study the changes in gene expression that occur over the course of the vascularization process, it became clear that a new approach to microarray experimental design and analysis was required. It was also clear that most researchers were ill-equipped to process and interpret the tens of thousands of data points generated by microarray experiments. To address these needs, a software package called CARMA (Computational Analysis of Replicated Measurements for Arrays) was developed to perform an analysis of variance (ANOVA) on microarray experiments that incorporate replicated measurements. Utilizing replicated measurement-based designs makes it possible to incorporate multiple samples into the experimental design and calculate both the magnitude and the statistical significance of the differences in gene expression between samples. Software was also developed to implement and compare different algorithms and distance metrics for performing hierarchical clustering. Hierarchical clustering groups genes together based on the similarity of their expression profiles, and is used to reduce the complexity of a microarray dataset and identify genes that may be involved in the same or related processes or under similar types of transcriptional control. Utilizing simulated datasets containing known clusters of genes, the ability of each each algorithm/distance metric combination to recover the original clusters was evaluated. Lastly, both CARMA and hierarchical clustering were utilized to analyze changes in gene expression during the process of vascularization in an experimental model of angiogenesis and blood vessel maturation. Based on high-level patterns of gene expression and morphological measurements obtained using this model, a multi-phase model of angiogenesis-based vascularization is presented consisting of an initial angiogenic phase, followed by a maturation and network remodeling phase.

Colorectal cancer (CRC) is the third leading cause of cancer related deaths. Rodent models of CRC are useful for evaluating diagnostic tools, therapeutics, and disease progression; however, an appropriate imaging tool is needed. Optical coherence tomography (OCT) is a non-destructive imaging modality readily packaged into small diameter endoscopes. Using a near- infrared light source, structural images are generated from index of refraction mismatches with resolutions of 2-15 mm at imaging depths of up to 1.3 mm. In contrast, laser-induced fluorescence (LIF) spectroscopy provides information about biochemical composition, exciting tissues with ultraviolet to green wavelengths of light to measure fluorescence emission from endogenous fluorophores such as NADH, collagen, and porphyrin.We apply OCT and LIF to mouse models of CRC, beginning with a comprehensive ex-vivo evaluation of normal mouse gastrointestinal (GI) tract in various strains and ages and secondarily sampled colorectal neoplasia and inflammatory bowel disease (IBD) using a combined in-air OCT/LIF system. A set of characteristic features of OCT images were developed for normal esophagus, small intestine, and colon; preliminary image feature criteria were also developed for colorectal neoplasia and IBD. LIF characterized the endogenous fluorescence of mouse GI tract, with spectral features corresponding to collagen, NADH, and hemoglobin. In the IBD sample, LIF emission displayed potentially diagnostic peaks at 635 and 670 nm, consistent with increased porphyrin production by bacteria associated with IBD.Next, endoscopic OCT/LIF was evaluated in an in-vivo serial study using a prototype 2 mm diameter endoscope to image the lower colon of ApcMin and control mice. Adenoma development over OCT imaging timepoints was characterized as a progressive mucosal thickening to frank mass formation. LIF spectral comparisons revealed decreased 405 nm intensity and the presence of a peak at 680 nm over adenoma.In a final study, ultrahigh resolution OCT (UHR OCT) was used to serially image the lower colon of azoxymethane treated A/J mice to monitor CRC progression and determine OCT's capability of identifying early disease. A panel of blinded mouse colon pathology experts assigned a diagnosis based on the OCT images, which was then compared to a histological diagnosis assigned by a blinded pathologist. At the final imaging timepoint, 95% of adenomas and 23% of gastrointestinal intraepithelial neoplasia (GIN, 38% protruding GIN and 9% non-protruding GIN) were correctly diagnosed. The panel identified 68% of disease foci (95% adenoma, 76% protruding GIN, and 13% non-protruding GIN). Over the OCT imaging timepoints, disease progression followed a typical succession, with normal or GIN preceding adenoma. Endoscopic UHR OCT enabled accurate diagnosis of adenomas, identification of protruding GIN, and non-destructive visualization of CRC progression, providing a tool for cancer research in animal models.