<p>This dissertation investigates the feasibility of a real-time transthoracic Acoustic Radiation Force Impulse (ARFI) imaging system to measure myocardial function non-invasively in clinical setting. Heart failure is an important cardiovascular disease and contributes to the leading cause of death for developed countries. Patients exhibiting heart failure with a low left ventricular ejection fraction (LVEF) can often be identified by clinicians, but patients with preserved LVEF might be undetected if they do not exhibit other signs and symptoms of heart failure. These cases motivate development of transthoracic ARFI imaging to aid the early diagnosis of the structural and functional heart abnormalities leading to heart failure.</p><p>M-Mode ARFI imaging utilizes ultrasonic radiation force to displace tissue several micrometers in the direction of wave propagation. Conventional ultrasound tracks the response of the tissue to the force. This measurement is repeated rapidly at a location through the cardiac cycle, measuring timing and relative changes in myocardial stiffness. ARFI imaging was previously shown capable of measuring myocardial properties and function via invasive open-chest and intracardiac approaches.</p><p>The prototype imaging system described in this dissertation is capable of rapid acquisition, processing, and display of ARFI images and shear wave elasticity imaging (SWEI) movies. Also presented is a rigorous safety analysis, including finite element method (FEM) simulations of tissue heating, hydrophone intensity and mechanical index (MI) measurements, and thermocouple transducer face heating measurements. For the pulse sequences used in later animal and clinical studies, results from the safety analysis indicates that transthoracic ARFI imaging can be safely applied at rates and levels realizable on the prototype ARFI imaging system. </p><p>Preliminary data are presented from <italic>in vivo</italic> trials studying changes in myocardial stiffness occurring under normal and abnormal heart function. Presented is the first use of transthoracic ARFI imaging in a serial study of heart failure in a <italic>porcine</italic> model. Results demonstrate the ability of transthoracic ARFI to image cyclically-varying stiffness changes in healthy and infarcted myocardium under good B-mode imaging conditions at depths in the range of 3-5 cm. Challenging imaging scenarios such as deep regions of interest, vigorous lateral motion and stable, reverberant clutter are analyzed and discussed.</p><p>Results are then presented from the first study of clinical feasibility of transthoracic cardiac ARFI imaging. At the Duke University Medical Center, healthy volunteers and patients having magnetic resonance imaging-confirmed apical infarcts were enrolled for the study. The number of patients who met the inclusion criteria in this preliminary clinical trial was low, but results showed that the limitations seen in animal studies were not overcome by allowing transmit power levels to exceed the FDA mechanical index (MI) limit. The results suggested the primary source of image degradation was clutter rather than lack of radiation force. Additionally, the transthoracic method applied in its present form was not shown capable of tracking propagating ARFI-induced shear waves in the myocardium.</p><p>Under current instrumentation and processing methods, results of these studies support feasibility for transthoracic ARFI in high-quality B-Mode imaging conditions. Transthoracic ARFI was not shown sensitive to infarct or to tracking heart failure in the presence of clutter and signal decorrelation. This work does provide evidence that transthoracic ARFI imaging is a safe non-invasive tool, but clinical efficacy as a diagnostic tool will need to be addressed by further development to overcome current challenges and increase robustness to sources of image degradation.</p> / Dissertation

<p>X-ray micro-CT is widely used for small animal imaging in preclinical studies of cardiopulmonary disease, but further development is needed to improve spatial resolution, temporal resolution, and material contrast. This study presents a set of tools that achieve these improvements. These tools include the mathematical formulation and computational implementation of algorithms for calibration, image reconstruction, and image analysis with our custom micro-CT system. These tools are tested in simulations and in experiments with live animals. With these tools, it is possible to visualize the distribution of a contrast agent throughout the body of a mouse as it changes over time, and produce 5-dimensional images (3 spatial dimensions + time + energy) of the cardiac cycle.</p> / Dissertation

<p>The redistribution of incident light into scattered fields ultimately limits the ability to image into biological media. However, these scattered fields also contain information about the structure and distribution of protein complexes, organelles, cells and whole tissues that can be used to assess the health of tissue or to enhance imaging contrast by excluding confounding signals. The interpretation of scattered fields depends on a detailed understanding of the scattering process as well as sophisticated measurement systems. In this work, the development of new instruments based on low coherence interferometry (LCI) is presented in order to perform precise, depth-resolved measurements of scattered fields. Combined with LCI, the application of new light scattering models based on both analytic and numerical methods is presented in order to interpret scattered field measurements in terms of scatterer geometry and tissue health. </p><p>The first portion of this work discusses the application of a new light scattering model to the measurement recorded with an existing technique, Angle Resolved Low Coherence Interferometry (a/LCI). In the a/LCI technique, biological samples are interrogated with collimated light and the energy per scattering angle at each depth in the volume is recorded interferometrically. A light scattering model is then used to invert the scattering measurements and measure the geometry of cell nuclei. A new light scattering model is presented that can recover information about the size, refractive index, and for the first time, shape of cell nuclei. This model is validated and then applied to the study of cell biology in a series of experiments measuring cell swelling, cell deformation, and finally detecting the onset of apoptosis.</p><p>The second portion of this work introduces an improved version of a/LCI based on two dimension angle resolved measurement (2D a/LCI) and Fourier domain low coherence interferometry (FD-LCI). Several systems are presenting for high speed and polarization-resolved measurements of scattered fields. An improved light scattering model based on fully polarization and solid angle resolved measurements is presented, and then efficiently implemented using distributed computing techniques. The combined system is validated with phantoms and is shown to be able to uniquely determine the size and shape of scattering particles using a single measurement.</p><p>The third portion of this work develops the use of angle-resolved interferometry for imaging through highly scattering media by exploiting the tendency of scatterers to forward scatter light. A new interferometers is developed that can image through very large numbers of scattering events with acceptable resolution. A computational model capable of reproducing experimental measurements is developed and used to understand the performance of the technique.</p><p>The final portion of the work develops a method for processing 2D angle resolved measurements using optical autocorrelation. In this method, measurements over a range of angles are fused into a single depth scan that incorporates the component of scattered light only from certain spatial scales. The utility of the technique is demonstrated using a gene knockout model of retinal degeneration in mice. Optical autocorrelation is shown to be a potentially useful biomarker of tissue disease.</p> / Dissertation

<p>Imaging based on nonlinear processes takes advantage of the localized excitation to achieve high spatial resolution, optical sectioning, and deeper penetration in highly scattering media. However, the use of nonlinear contrast for imaging has conventionally been limited to processes that create light of wavelengths that are different from the wavelengths used for excitation. Intrinsic nonlinear contrasts that do not generate light at distinct wavelengths are generally difficult to measure because of the overwhelming background from the excitation light. This dissertation focuses on extension of nonlinear microscopy to these new intrinsic processes by using femtosecond pulse shaping to encode the nonlinear information as new frequency components in the spectrum. We will present a pump-probe microscopy technique based on pulse train shaping technology to sensitively access nonlinear transient absorption or gain processes. This technique has recently been used to uniquely identify a variety of biological pigments with high spatial resolution. Here, we extend this technique to image and characterize several inorganic and organic pigments used in historical artwork. We also present a spectral reshaping technique based on individual femtosecond pulse shaping to sensitively access nonlinear refractive contrasts in scattering media. We will describe an extension of this technique to utilize two distinct wavelengths and discuss its application in biological imaging. This two-color implementation would allow the extension of widely employed phase contrast to the nonlinear regime.</p> / Dissertation

<p>Positron emission tomography (PET) is a nuclear medicine diagnostic imaging exam of metabolic processes in the body. Radiotracers, which consist of positron emitting radioisotopes and a molecular probe, are introduced into the body, emitted radiation is detected, and tomographic images are reconstructed. The primary clinical PET application is in oncology using a glucose analogue radiotracer, which is avidly taken up by some cancers.</p><p>It is well known that PET performance and image quality degrade as body size increases, and epidemiological studies over the past two decades show that the adult US population's body size has increased dramatically and continues to increase. Larger patients have more attenuating material that increases the number of emitted photons that are scattered or absorbed within the body. Thus, for a fixed amount of injected radioactivity and acquisition duration, the number of measured true coincidence events will decrease, and the background fractions will increase. Another size-related factor, independent of attenuation, is the volume throughout which the measured coincidence counts are distributed: for a fixed acquisition duration, as the body size increases, the counts are distributed over a larger area. This is true for both a fixed amount of radioactivity, where the concentration decreases as size increases, and a fixed concentration, where the amount radioactivity increases with size.</p><p>Time-of-flight (TOF) PET is a recently commercialized technology that allows the localization, with a certain degree of error, of a positron annihilation using timing differences in the detection of coincidence photons. Both heuristic and analytical evaluations predict that TOF PET will have improved performance and image quality compared to non-TOF PET, and this improvement increases as body size increases. The goal of this dissertation is to parameterize the image quality improvement of TOF PET compared to non-TOF PET as a function of body size. Currently, no standard for comparison exists.</p><p>Previous evaluations of TOF PET's improvement have been made with either computer-simulated data or acquired data using a few discrete phantom sizes. A phantom that represents a range of attenuating dimensions, that can have a varying radioactivity distribution, and that can have radioactive inserts positioned throughout its volume would facilitate characterizing PET system performance and image quality as a function of body size. A fillable, tapered phantom, was designed, simulated, and constructed. The phantom has an oval cross-section ranging from 38.5 &times; 49.5 cm to 6.8 &times; 17.8 cm, a length of 51.1 cm, a mass of 6 kg (empty), a mass of 42 kg (water filled), and 1.25-cm acrylic walls.</p><p>For this dissertation research, PET image quality was measured using multiple, small spheres with diameters near the spatial resolution of clinical whole-body PET systems. Measurements made on a small sphere, which typically include a small number of image voxels, are susceptible to fluctuations over the few voxels, so using multiple spheres improves the statistical power of the measurements that, in turn, reduces the influence of these fluctuations. These spheres were arranged in an array and mounted throughout the tapered phantom's volume to objectively measure image quality as a function of body size. Image quality is measured by placing regions of interest on images and calculating contrast recovery, background variability, and signal to noise ratio.</p><p>Image quality as a function of body size was parameterized for TOF compared to non-TOF PET using 46 1.0-cm spheres positioned in six different body sizes in a fillable, tapered phantom. When the TOF and non-TOF PET images were reconstructed for matched contrast, the square of the ratio of the images' signal-to-noise ratios for TOF to non-TOF PET was plotted as a function, <italic>f</italic>(<italic>D</italic>), of the radioactivity distribution size, <italic>D</italic>, in cm. A linear regression was fit to the data: <italic>f</italic>(<italic>D</italic>) = 0.108<italic>D</italic> - 1.36. This was compared to the ratio of <italic>D</italic> and the localization error, <italic>&sigma;_d</italic>, based on the system timing resolution, which is approximately 650 ps for the TOF PET system used for this research. With the image quality metrics used in this work, the ratio of TOF to non-TOF PET fits well to a linear relationship and is parallel to <italic>D/&sigma;_d</italic>. For <italic>D</italic> < 20 cm, there is no image quality improvement, but for radioactivity distributions <italic>D</italic> > 20 cm, TOF PET improves image quality over non-TOF PET. PET imaging's clinical use has increased over the past decade, and TOF PET's image quality improvement for large patients makes TOF an important new technology because the occurrence of obesity in the US adult population continues to increase.</p> / Dissertation

<p>This dissertation aims to demonstrate the feasibility of direct infarct visualization using 3D medical ultrasound. The dissertation proceeds by providing the first ever demonstration of fully-sampled 3D ultrasonic speckle tracking using raw B-Mode data of the heart. The initial demonstration uses a Cramer-Rao lower bound limited displacement estimator. The dissertation then proceeds to develop an implementable method for biased time-delay estimation. Biased time-delay estimation is shown to surpass the traditional limits described by the Cramer-Rao lower bound in a mean square error sense. Additional characterization of this new class of estimator is performed to demonstrate that with easily obtainable levels of prior information it is possible to estimate displacements that do surpass the Cramer-Rao lower bound. Finally, using 2D and 3D realizations of biased displacement estimation (Bayesian speckle tracking) the passive strain induced in the ventricle walls during atrial systole is shown to be sufficient to distinguish healthy and chronically infarcted myocardium.</p> / Dissertation

We have designed and demonstrated a new class of medical navigation methods that use the fingerprint-like biometrically distinct ultrasound echo patterns produced by different locations in tissue. As an example of this new biometric navigation approach, we have constructed and tested a system that uses ultrasound data to achieve prospective motion compensation in MRI, especially for respiratory motion during interventional MRI procedures in moving organs such as the liver. The ultrasound measurements are collated with geometrical information from MRI during a training stage to form a mapping table that relates ultrasound measurements to positions. During prospective correction, the system makes frequent ultrasound measurements and uses the map to determine the corresponding position. Results in motorized linear motion phantoms and freely breathing animals indicate that the system performs well. Apparent motion is reduced by up to 97.8%, and motion artifacts are reduced or eliminated in 2D Spoiled Gradient-Echo images. The motion compensation is sufficient to permit MRI thermometry of focused ultrasound heating during respiratory-like motion, with results similar to those obtained in the absence of motion. This new technique may have applications for MRI thermometry and other dynamic imaging in the abdomen during free breathing. We have also extended this technique to situations in which external position information during training is unavailable or incomplete, by extending the concept of Simultaneous Localization and Mapping to include determining the topology of a dense motion path through a gaussian random field. In the course of these investigations, we have also developed modified forms of referenceless MRI thermometry and Kalman filtering, specially adapted to optimize accuracy under our experimental conditions.

Fluorescence diffuse optical tomography provides non-invasive, in vivo imaging of molecular targets in small animals. While standard fluorescence microscopy is limited to shallow depths and small fields of view, tomographic methods allows recovery of the distribution of fluorescent probes throughout the small animal body. In this thesis, we present novel reconstruction algorithms for the tomographic separation of optical parameters using time-domain (TD) measurements. These technique are validated using simulations and with experimental phantom and mouse imaging studies. We outline the contributions of each chapter of the thesis below. First, we explore the TD fluorescence tomography reconstruction problem for single and multiple fluorophores with discrete lifetimes. We focus on late arriving photons and compare a direct inversion approach with a two-step, asymptotic approach operating on the same TD data. We show that for lifetime multiplexing, the two methods produce fundamentally different kinds of solutions. The direct inversion is computationally inefficient and results in poor separation but has overall higher resolution while the asymptotic approach provides better separation, relative quantitation of lifetime components and localization but has overall lower resolution. We verify these results with simulation and experimental phantoms. Second, we introduce novel high resolution lifetime multiplexing algorithms which combine asymptotic methods for separation of fluorophores with the high resolving power of early photon tomography. We show the effectiveness of such methods to achieve high resolution reconstructions of multiple fluorophores in simulations with complex-shaped phantoms, a digital mouse atlas and also experimentally in fluorescent tube phantoms. Third, we compare the performance of tomographic spectral and lifetime multiplexing. We show that both of these techniques involve a two-step procedure, consisting of a diffuse propagation step and a basis-function mixing step. However, in these two techniques, the order of the two steps is switched, which leads to a fundamental difference in imaging performance. As an illustration of this difference, we show that the relative concentrations of three colocalized fluorophores in a diffuse medium can accurately be retrieved with lifetime methods but cannot be retrieved with spectral methods. Fourth, we address the long standing challenge in diffuse optical tomography (DOT) of cross-talk between absorption and scattering. We extend the ideas developed from lifetime multiplexing algorithms by using a constrained optimization approach for separation of absorption and scattering in DOT. Using custom designed phantoms, we demonstrate a novel technique allows better separation of absorption and scattering inclusions compared to existing algorithms for CW and TD diffuse optical tomography. Finally, we show experimental validation of the lifetime multiplexing algorithms developed in this thesis using three experimental models. First, we show the reconstruction of overlapping complex shapes in a dish phantom. Second, we demonstrate the localization accuracy of lifetime based methods using fluorescent pellets embedded in a sacrificed mouse. Third, we show using planar imaging and tomography, the in vivo recovery of multiple anatomically targeted near-infrared fluorophores. In summary, we have presented novel reconstruction algorithms and experimental methods that extend the capability of time-domain fluorescence diffuse optical tomography systems. The methods developed in this thesis should also have applicability for general multi-parameter image reconstruction problems. / Engineering and Applied Sciences

As new therapeutic targets and drugs are discovered for B-cell lymphoma and other cancers, companion diagnostics are also needed to determine target engagement, therapeutic efficacy, and patient segmentation for clinical trials. We first employed synthetic chemistry to build a platform for modifying small molecule drugs into imaging probes, using the poly(ADP-ribose) polymerase 1 (PARP1) inhibitor AZD2281 (Olaparib) as a model for technology development. Our results showed that small-molecule companion imaging drugs can be used for fluorescence imaging in cells, as well as for pharmacokinetic studies and positron emission tomography (PET) imaging in vivo, without significantly perturbing their target binding properties or cellular uptake. To apply this approach to B-cell lymphoma drugs currently in clinical trials, we modified an irreversible inhibitor of Bruton's Tyrosine Kinase (BTK), PCI-32765 (Ibrutinib), with the fluorophore Bodipy FL (BFL), and used it for imaging in cells and in a mouse window-chamber xenograft model. The excellent co-localization of our probe (Ibrutinib-BFL) with BTK demonstrated its utility for studying additional BTK inhibitors and as a companion imaging probe. In parallel, we hypothesized that central nervous system (CNS) lymphoma diagnosis from paucicellular cerebrospinal fluid (CSF) samples could be improved with molecular profiling of putative lymphoma cells trapped in a customized microfluidic chip. Following fabrication and characterization of a polydimethylsiloxane (PDMS) diagnostic device containing an array of affinity-free single-cell capture sites, we were able to efficiently recover >90% of lymphocytes, perform immunostaining on chip, and apply an image-processing algorithm to group cells based on their molecular marker expression, such as kappa/lambda light chain restriction. Additionally, in combination with Ibrutinib-BFL or other imaging drugs, we demonstrated the potential for on-chip drug imaging for use in conjunction with drug development. Finally, we applied bioorthogonal conjugation chemistries on cellulose paper for potential applications in lowering the cost of drug screening. We anticipate that these approaches will enable direct, molecular information for personalized treatment decisions in B-cell lymphomas, as well as provide a roadmap for the development of companion diagnostic probes and devices for additional indications.

<p>Tree-like structures are fundamental in nature. A wide variety of two-dimensional imaging techniques allow us to image trees. However, an image of a tree typically includes spurious branch crossings and the original relationships of ancestry among edges may be lost. We present a methodology for estimating the most likely topology of a rooted, directed, three-dimensional tree given a single two-dimensional image of it. We regularize this inverse problem via a prior parametric tree-growth model that realistically captures the morphology of a wide variety of trees. We show that the problem of estimating the optimal tree has linear complexity if ancestry is known, but is NP-hard if it is lost. For the latter case, we present both a greedy approximation algorithm and a heuristic search algorithm that effectively explore the space of possible trees. Experimental results on retinal vessel, plant root, and synthetic tree datasets show that our methodology is both accurate and efficient.</p> / Dissertation