Browsing by Author "Warren, Warren S"

Recent work has shown that hyperpolarized magnetic resonance spectroscopy (HP-MRS) can trace in vivo metabolism of biomolecules and is therefore extremely promising for diagnostic imaging. The most severe challenge this technique faces is the short signal lifetime for hyperpolarization, which is dictated by the spin-lattice (T1) relaxation. In this thesis we show with theory, simulation and experiment that the long-lived nuclear spin states in chemically equivalent or near equivalent spin systems offer a solution to this problem. Spin polarization that has lifetime much longer than T1 (up to 70-fold) has been demonstrated with pulse sequence techniques that are compatible with clinical imaging settings. Multiple classes of molecules have been demonstrated to sustain such long-lived hyperpolarization.

Any type of microscopy faces the problem in an attenuated signal level and reduced optical resolution due to optical aberration. To overcome this problem, adaptive optics was implemented in a two-photon fluorescence microscope. Using a “sensorless” approach, this study corrected the aberration in the system using two different excitation colors. As methodology, the point spread function was compared before and after applying adaptive optics. This study demonstrated that adaptive optics improves the resolution of the microscope for both excitation wavelengths. The aberration correction was then monitored as a function of depth. The result showed an improvement in the optimization metric as imaging depth is increased. Thus, adaptive optics offers improved imaging of the sample at deeper depths with better optical resolution and higher signal-to-noise ratio.

The explosion in the popularity and success of deep learning in the past decade has accelerated the development of computationally efficient, GPU-accelerated frameworks, such as TensorFlow and PyTorch, for rapid prototyping of neural networks. In this dissertation, we show that these deep learning tools are also well-suited for computational 3D imaging problems, specifically optical diffraction tomography (ODT), photogrammetry, and our newly proposed optical coherence refraction tomography (OCRT). Underlying these computational 3D imaging techniques is a physical model that demands multiple measurements taken with either angular diversity, wavelength diversity, or both. This requirement can be compactly summarized as wavevector (or k-vector) diversity, where the magnitude and direction of the wavevector correspond to the color and angle of the light, respectively.

To understand the importance of wavevector diversity for 3D imaging, this dissertation starts by advancing a unified k-space theory of optical coherence tomography (OCT), the most comprehensive and inclusive theoretical description of OCT to date that not only describes the transfer functions of all major forms of OCT and other coherent techniques (e.g., confocal microscopy, holography, ODT), but also includes the fundamental concepts of OCT, such as speckle, dispersion, aberration, and the tradeoff between lateral resolution and depth of focus (DOF).

Consistent with this unified theory, we implemented in TensorFlow a reconstruction algorithm for ODT, a technique that relies on illumination angular diversity to achieve 3D refractive index (RI) imaging. We propose a new method for filling the well-known “missing cone” of the ODT transfer function by reparameterizing the 3D sample as the output of an untrained neural network known as a deep image prior (DIP), which we show to outperform traditional regularization strategies.

Next, we introduce OCRT, a computational extension of OCT that incorporates extreme angular diversity over OCT's already high wavelength diversity to enable resolution-enhanced, speckle-reduced reconstructions that overcome the lateral-resolution-DOF tradeoff. OCRT also jointly reconstructs quantitative RI maps of the sample using a ray-based physical model implemented in TensorFlow. We also demonstrate spectroscopic OCRT (SOCRT), an extension of spectroscopic OCT (SOCT) that overcomes its tradeoff between spectral and axial resolution.

Motivated to make OCRT more widely applicable, we propose a new use of conic-section (e.g., parabolic, ellipsoidal) mirrors to allow fast multi-view imaging over very high angular ranges (up to 360°) using galvanometers without requiring sample rotation. We theoretically characterize the achievable fields of view (FOVs) as a function of many imaging system parameters (e.g., NA, wavelength, incidence angle, focal length, and telecentricity). Based on these predictions, we constructed a parabolic-mirror-based imaging system that facilitates multi-view OCT volume capture with millimetric FOVs over up to ±75°, which we combined to perform 3D OCRT reconstructions of zebrafish, fruitfly, and mouse tissue.

Finally, we adapted the OCRT reconstruction algorithm to photogrammetric 3D mesoscopic imaging with tens-of-micron accuracy, using a sequence of smartphone camera images taken at close range under freehand motion. 3D estimation was possible due to the angular diversity afforded by the nontelecentricity of smartphone cameras, using a similar ray-based model as for OCRT. We show that careful modeling of lens distortion and incorporation of a DIP are both pivotal for obtaining high 3D accuracy using devices not designed for close-range imaging.

This dissertation reports advances in magnetic resonance imaging (MRI), with the ultimate goal of improving signal and contrast in biomedical applications. More specifically, novel MRI pulse sequences have been designed to characterize microstructure, enhance signal and contrast in tissue, and image functional processes. Using these pulse sequences, intermolecular multiple quantum coherence (iMQC) signals that arise from the dipolar field over well-defined distances can be observed; these signals were used here to probe material microstructure. Using iMQCs, the restricted diffusion in uni- and multi-lamellar vesicles such as liposomes and polymersomes was characterized, with potential applications for monitoring drug transport and release; moreover, mesoscopic anisotropy in developing rat brains was studied, which required significant pulse sequence optimizations and corrections to the original dipolar field framework. We have also developed and applied modified multipulse echo sequences with optimized interpulse delays for tissue imaging. These sequences have enhanced the signal and may provide new contrast in various tissues, including normal, tumor, and fatty tissues. Finally, the use of MRI to study functional processes, including temperature and perfusion, is described.

The purpose of this thesis is to report technological developments in contrast mechanisms for MRI. The search for new forms of contrast is on-going, with the hope that new contrast mechanisms and new contrast agents will provide unique insights into various molecular processes and disease states. In this thesis, we will describe new contrast mechanisms developed by manipulating the inherent physics of the system, as well as the development of exogenous contrast agents. More specifically, we will describe the application of iMQCs (intermolecular multiple quantum coherences) to thermometry and structural imaging, and the unique information provided from these studies. We will also describe methods for migrating iMQC-based pulse sequences from a Bruker research console onto a clinical GE console, thus enabling the application of iMQCs to humans. We will describe the development of hyperpolarized contrast agents which have the potential to provide an unprecedented level of molecular contrast to MRI and the development of techniques to enhance the lifetime of these hyperpolarized contrast agents. Finally, we will discuss a new type of T2 -weighted imaging which significantly improves the refocusing of CPMG-type sequences.

Nonlinear optical microscopies provide a flexible, inexpensive, and useful way to do medical imaging. Among the imaging modalities, pump-probe microscopy is a versatile technique that shows great potential to image melanomas (a skin cancer) for which there is still a great need for improved early clinical diagnosis. But image analysis often suffers from an incomplete understanding of the signal properties. In this dissertation, I discuss several unusual properties observed in pump-probe signals using model samples (e.g., gold nanoparticles, natural and synthesized melanin particles, historic artwork pigments) and an upgraded pump-probe microscope. The experiments improve understanding in pump-probe signals and help for better data analysis.Chapter 2 demonstrates my experimental results regarding the directional dependence of pump-probe signals. I upgrade imaging system for angle-resolved measurements, and use melanin particles and gold nanoparticles to study the role of scattering in the generation of pump-probe signals. The results show the reason for signal discrepancy between transmission (for thin sliced samples) and backward (for in-vivo imaging) detections observed in melanoma imaging. Chapter 3 discusses an intrinsic polarization anisotropy in the pump-probe signal of melanin. I develop a formalism based on transition dipole moments to model this polarization anisotropy, and use it to decompose pump-probe signals in several pigments (melanin, historic artwork pigments). I improve the imaging system to accurately control and calibrate the polarization angle of the lasers for polarization measurements. The polarization angle serves as a useful imaging parameter. Combined with the directional dependence discussed in chapter 2, a global fitting is performed on the melanin signals to extract strengths and decays of the various components in the signal. Chapter 4 discusses studies on remaining puzzles observed in melanoma imaging, including depth dependence, signal dynamics change due to particle aggregation, and a long-lived signal component of unclear origin. While these anomalies are not fully understood, their properties are presented for future studies. Chapter 5 summarizes the key achievements presented in this dissertation, and discusses new questions that should be explored in the future.

For women with a high risk of developing breast cancer, the treatment options can be drastic, unnecessary, and emotionally and physically scarring. While histological evaluation of tissue samples is reliable for invasive breast cancers, the malignant potential from ductal carcinoma in situ (DCIS) cannot be reliably predicted from histopathology. Biomarkers in the breast have appeal as being directly reflective of underlying preneoplastic processes and could be used to accurately predict short-term breast cancer risk even in DCIS. By using 1H NMR metabonomics on interstitial breast fluid provided from random periareaolar fine-needle aspirations, as well as wash fluid from breast extractions, some potential biomarkers have been uncovered that could aid physicians in the prognosis of high-risk women.

Femtosecond pump-probe microscopy is a specific implementation of nonlinear microscopy that can measure transient absorption processes like two-photon absorption, excited state absorption, stimulated emission, ground state depletion, and stimulated Raman scattering for structural and functional imaging. Transient absorption can provide molecular specificity without the need for exogenous dyes and labels, and measure the electronic and vibrational dynamics of materials with high specificity. Because of its good sensitivity, pump-probe microscopy has found applications in the fields of material science, biomedicine, and art conservation. For example, in biomedicine, our group (Warren lab at Duke University) have shown that the characteristic pump-probe signature of melanin changes with increased metastatic potential, which can directly impact patient treatment and prognosis by assisting pathologists in the diagnostic process.

However, current pump-probe microscopy suffers from several limitations. Firstly, the system is bulky and not user-friendly, which makes clinical translation difficult. Secondly, the pump-probe signal is in general overlaid with parasitic signals stemming from fluorescence and thermal lensing. The spurious fluorescence and thermal background not only limit the detection of molecules with weak pump-probe response, but also lead to inaccurate interpretation of signal magnitude in quantitative data analysis. Finally, the field-of-view, working-distance and penetration depth of pump-probe imaging needs to be improved for broader ranges of applications.

In this work, I address these issues by exploiting a broadband laser source, as well as new temporal and spatial configurations of pump and probe beams. In particular, in Chapter 2, I discuss a compact and user-friendly pump-probe system with high temporal resolution, based on a single, ultra-broadband, femtosecond laser source. In Chapter 3, I propose a novel detection method, featuring a MHz time-delay, polarization, and pulse-width modulation scheme to suppress unwanted signals and enhance the nonlinear imaging contrast. Finally, in Chapters 4 and 5, I propose a crossed-beam method for both Gaussian and Airy beams and demonstrate their promise for depth-resolved, long working-distance, large field-of-view imaging in scattering media. My work thus overcomes many of the issues in current instrumentation and paves the way for the next-generation femtosecond pump-probe microscopy.

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.

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.

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.

First mined in Afghanistan nearly 6,000 years ago, lapis lazuli is a blue pigment also known as ultramarine, a material that was highly prized (and corresponding highly priced) by Western painters during the Middle Ages and the Italian Renaissance. Both the mineral lapis lazuli and the synthetic ultramarine can undergo degradation in paintings and other works of cultural heritage, which presents challenges to the preservations of these works. Due to the limitations of many modern analytical techniques, art conservators and conservation scientists often still need to remove a sample of paint in order to understand the layering of pigments in a painting. Femtosecond transient absorption (also called pump-probe) spectroscopy and imaging are here used to explore the effects of depth, polarization, and power on the ultrafast excited state dynamics of ultramarines both natural and synthetic, in order to further understand the photophysics and potential photo-degradation pathways of ultramarine pigments in paintings. Both lapis lazuli and synthetic ultramarine undergo identical forms of photo-induced transformation in the context of these experiments, where it appears that either the lazurite chromophore or the sodalite cage structure of ultramarine is destroyed.

Two-photon excited fluorescence microscopy (TPM) has become a standard biological imaging tool due to its simplicity and versatility. The fundamental contrast mechanism is derived from fluorescence of intrinsic or extrinsic markers via simultaneous two-photon absorption which provides inherent optical sectioning capabilities. The NIR-II wavelength window (1000–1350 nm), a new biological imaging window, is promising for TPM because tissue components scatter and absorb less at longer wavelengths, resulting in deeper imaging depths and better contrasts, compared to the conventional NIR-I imaging window (700–1000 nm). However, the further enhancement of TPM has been hindered by a lack of good two-photon fluorescent imaging markers in the NIR-II.

In this dissertation, we design and characterize novel two-photon imaging markers, optimized for NIR-II excitation. More specifically, the work in this dissertation includes the investigation of two-photon excited fluorescence of various highly conjugated porphyrin arrays in the NIR-II excitation window and the utilization of nanoscale polymersomes that disperse these highly conjugated porphyrin arrays in their hydrophobic layer in aqueous environment. The NIR-emissive polymersomes, highly conjugated porphyrins-dispersed polymersomes, possess superb two-photon excited brightness. The synthetic nature of polymersomes enables us to formulate fully biodegradable, non-toxic and surface-functionalized polymersomes of varying diameters, making them a promising and fully customizable multimodal diagnostic nano-structured soft-material for deep tissue imaging at high resolutions. We demonstrated key proof-of-principle experiments using NIR-emissive polymersomes for in vivo two-photon excited fluorescence imaging in mice, allowing visualization of blood vessel structure and identification of localized tumor tissue. In addition to spectroscopic characterization of the two-photon imaging agents and their imaging capabilities/applications, the effect of the laser setup (e.g., repetition rate of the laser, peak intensity, system geometry) on two-photon excited fluorescence measurements is explored to accurately measure two-photon absorption (TPA) cross-sections. A simple pulse train shaping technique is demonstrated to separate pure nonlinear processes from linear background signals, which hinders accurate quantification of TPA cross-sections.

There are two fundamental limitations in magnetic resonance: the poor signal amplitude and the short duration before the system return to equilibrium. Hyperpolarization methods solve the problem of signal amplitude, however, the duration of the hyperpolarized signal is still limited by the spin-lattice relaxation time, T1. Disconnected eigenstates provide a mechanism by which hyperpolarization can be stored for several times T1. This thesis contributes to the knowledge of these states in four important ways. First, the decay of hyperpolarized magnetization of gas is simulated in lung tissue with a contrast agent, yielding insights about the optimal field strength for imaging. Second, I show that it is possible to rapidly discover and characterize disconnected eigenstates by showing that they can be measured without synthesizing the isotopically labeled compound. Third, I extend the spin systems that can support disconnected eigenstates by expanding the theory to include spin-1 nuclei. Finally, I show that disconnected states with long lifetimes can be populated in conjunction with hyperpolarization techniques to simultaneously yield large signal amplitudes for long durations.

Applications of hyperpolarized spin order are likely to be in complex geophysical or biological structures. Understanding the effect of the inhomogeneous fields created when such structures are placed in a magnetic field on hyperpolarized spin order is a necessity to characterize the experimental signal. An example case of hyperpolarized 3He and 129Xe diffusing through lung tissue is examined. In particular a Monte Carlo simulation tool, combined with a magnetic field map of the inhomogeneous field created by mouse lung tissue, is used to determine the dephasing rate of hyperpolarized 3He and 129Xe in the presence of SuperParamagnetic Iron Oxide Nanoparticles (SPION). Contributions to the dephasing rate include the inhomogeneous field, the SPION magnetic field, and dephasing caused by collisions with the confining geometry. The sensitivity of either gas to SPION increases with increasing SPION concentration and decreasing field strength.

There are some general rules about what makes for a disconnected eigenstate (or singlet state) with a long lifetime. However, no systematic experimental study has been undertaken due to the cost and time-constraints of synthesizing the labeled species for study. I show that synthesis is not a barrier for characterizing the long-lived states. Instead the lifetimes may be determined by using the naturally occurring doubly-labeled isotopomer. I verified this method with two compounds, diphenyl acetylene (DPA) and diethyl oxylate (DEO). The former was determined to have a singlet lifetime TS = 251.40 ±3.16 s from the synthesized species, while the naturally occurring isotopomer yielded a lifetime TS = 202 ±55.30 s, both substantially longer than the spin-lattice relaxation time, T1 = 1.63 ±0.01s. In DEO, the lifetime from the disconnected eigenstate was determined to be TS = 14.62 ±0.76 s (synthesized), TS = 19.32 ±3.16 s (naturally occurring). This method is applied to a range of compounds ranging from simple four-spin systems, such as diacetylene (TS = 48.80 ±22.74 s, T1 = 18.66 ±1.16 s) to eight spin systems in dimethylmaleic anhydride (TS = 27.25 ±3.39 s, T1 = 9.38 ±0.43 s). Additionally, a family of compounds including naphthalene (TS = 4.37 ±0.34 s, T1 = 11.33 ±4.89 s), biphenyl (TS = 3.09 ±0.66 s, T1 = 4.69 ±0.10 s), and DPA show that the rotation of the phenyl rings and intermolecular dipole-dipole relaxation can be critical to the relaxation dynamics.

One particular method of accessing the disconnected eigenstate involves coupling a chemically equivalent spin-1/2 pair asymmetrically to an auxiliary spin-1/2 pair. I demonstrate that the disconnected state may still be accessed when the auxiliary nuclei are spin-1. This has two distinct advantages. When the auxiliary nuclei change from proton to deuterium, the couplings are reduced by a factor of ~6.5 which prevents the disconnected state from relaxing as rapidly back to equilibrium. This is demonstrated in diacetylene-d2 and DPA-d10, where the singlet lifetime was extended by a factor of ~1.7 via deuteration (TS,1H = 49 ±23 s, TS,2H = 83 ±30 s for diacetylene and TS,1H = 274 ±6.1 s, TS,2H = 479 ±83 s for DPA). Additionally, by reducing the coupling strength, deuteration allows additional structural moieties to be explored, such as RDC=CDR. One such structure is explored in trans-ethylene-d2, where the singlet character of the protons can be accessed by the reduced coupling to the deuterium. Additionally, this allows for a relatively strong deuterium-deuterium scalar coupling, requiring modification to the theory. This is carried out analytically, and implications for the relaxation properties are performed using a spin-dynamics numerical simulation. The lifetime of the disconnected state was determined to be TS = 30.2 ±12.3 s, compared to the T1 = 1.1 ±0.2 s at high concentration (270 mM), and increasing to TS = 117. ±9.80 s at low concentration (52 mM). The variation in long lifetime is attributed to intermolecular dipole-dipole relaxation.

Ultimately, the gains in lifetime from using disconnected eigenstates provide a means to the practical implementation of hyperpolarization in a wider range of experiments. A recent hyperpolarization method, Signal Amplification By Reversible Exchange in Shield Enables Alignment Transfer to Heteronuclei (SABRE-SHEATH) is shown to directly hyperpolarize long lived spin order in a diazirine containing molecule. Diazirine rings are three member N=N-C groups that can replace a methylene group and serve as a versatile MR and optical molecular tag. Hyperpolarization is accomplished by bubbling parahydrogen through a solution containing the diazirine and an iridium catalyst. Due to the chemical inequivalence of the 15N of the diazirine, hyperpolarization of longitudinal magnetization and singlet character could be observed by transfer to the high field spectrometer. Signal enhancements of over 14,000 were observed. The magnetic field strength required for buildup of magnetization and singlet character was derived and is in agreement with the experiment. The magnetization lifetime was observed to be T1 = 5.75 ±0.18 minutes and independent of field strength, while the lifetime of the singlet character was observed to be as long as TS = 30.1 ±13.4 minutes at low field (3 Gauss).

The combination of these experiments – understanding lifetimes in inhomogeneous magnetic fields that will be encountered in experiment, identification of disconnected eigenstates with long lifetimes via the naturally occurring isotopomer and extending these lifetimes even further with deuteration, and finally, the direct generation of long-lived hyperpolarized spin order – allows a measurement that required hyperpolarized spin order for the enhanced signal amplitude, to be carried out.

Dermatopathologists need more reliable tools for analyzing biopsies of lesions that are potentially melanomas and determining the best treatment plan for the patient. Previously inaccessible, the chemical and physical properties of melanin provide insight into melanoma biochemistry. Two-color, near-infrared pump-probe microscopy of unstained, human pathology slides reveals differences in the type of melanins and the distribution of melanins between melanomas and benign nevi. Because the pump-probe response of melanin is resilient to aging, even for hundreds of millions of years, this tool could prove useful in retrospective studies to correlate melanin characteristics with patient outcome, thus eliminating the pathologist's uncertainty from the development of this classification method.

Pump-probe spectroscopy of a variety of melanin preparations including melanins with varying amounts of metal ions and toxins, those that have been photo-damaged or chemically oxidized, and melanins with a homogeneous size distribution shows that the pump-probe response is sensitive to these chemical and physical differences, not just melanin type as previously hypothesized. When sampling the response at several pump wavelengths, the specificity of this technique is derived from the absorption spectra of the underlying chromophores. Therefore, hyperspectral pump-probe microscopy of melanin could serve as an indicator of the chemical environment in a variety of biological contexts. For example, the melanin chemistry of macrophages suggests that these cells oxidize, homogenize, and compact melanin granules; whereas melanocytes produce heterogeneous melanins.

SABRE (Signal Amplification by Reversible Exchange) methods provide a simple, fast, and cost-effective method to hyperpolarize a wide variety of molecules in solution, and have been demonstrated with protons and, more recently, with heteronuclei (X-SABRE). In this dissertation, we first present oscillating pulse SABRE that use magnetic fields far away from the resonance condition of continuous excitation and can commonly triple the polarization. An analysis with average Hamiltonian theory indicates that the oscillating pulse, in effect, adjusts the J-couplings between hydrides and target nuclei and that a much weaker coupling produces maximum polarization. This theoretical treatment, combined with simulations and experiment, shows substantial magnetization improvements relative to traditional X-SABRE methods. It also shows that, in contrast to most pulse sequence applications, waveforms with reduced time symmetry in the toggling frame make magnetization generation more robust to experimental imperfections. A high-pressure SABRE approach is presented to enhance the exchange rate of the dihydride by increasing its concentration. To achieve this, two methods are proposed to improve the concentration of hydrogen gas: the brute-force high-pressure method and the supercritical SABRE method. The brute-force approach has been found to effectively increase the polarization by over three times. Further numerical analysis has shown that combining the oscillating pulse technique with the high-pressure method, and implementing temperature control, can effectively further enhance the polarization to higher levels.