Advanced Monte Carlo Simulation and Machine Learning for Frequency Domain Optical Coherence Tomography

Optical Coherence Tomography(OCT) is a popular, rapidly growing imaging technique with an increasing number of bio-medical applications due to its noninvasive nature. However, there are three major challenges in understanding and improving an OCT system: (1) Obtaining an OCT image is not easy. It either takes a real medical experiment or requires days of computer simulation. Without much data, it is difficult to study the physical processes underlying OCT imaging of different objects simply because there aren't many imaged objects. (2) Interpretation of an OCT image is also hard. This challenge is more profound than it appears. For instance, it would require a trained expert to tell from an OCT image of human skin whether there is a lesion or not. This is expensive in its own right, but even the expert cannot be sure about the exact size of the lesion or the width of the various skin layers. The take-away message is that analyzing an OCT image even from a high level would usually require a trained expert, and pixel-level interpretation is simply unrealistic. The reason is simple: we have OCT images but not their underlying ground-truth structure, so there is nothing to learn from. (3) The imaging depth of OCT is very limited (millimeter or sub-millimeter on human tissues). While OCT utilizes infrared light for illumination to stay noninvasive, the downside of this is that photons at such long wavelengths can only penetrate a limited depth into the tissue before getting back-scattered. To image a particular region of a tissue, photons first need to reach that region. As a result, OCT signals from deeper regions of the tissue are both weak (since few photons reached there) and distorted (due to multiple scatterings of the contributing photons). This fact alone makes OCT images very hard to interpret.

This thesis addresses the above challenges by successfully developing an advanced Monte Carlo simulation platform which is 10000 times faster than the state-of-the-art simulator in the literature, bringing down the simulation time from 360 hours to a single minute. This powerful simulation tool not only enables us to efficiently generate as many OCT images of objects with arbitrary structure and shape as we want on a common desktop computer, but it also provides us the underlying ground-truth of the simulated images at the same time because we dictate them at the beginning of the simulation. This is one of the key contributions of this thesis. What allows us to build such a powerful simulation tool includes a thorough understanding of the signal formation process, clever implementation of the importance sampling/photon splitting procedure, efficient use of a voxel-based mesh system in determining photon-mesh interception, and a parallel computation of different A-scans that consist a full OCT image, among other programming and mathematical tricks, which will be explained in detail later in the thesis.

Next we aim at the inverse problem: given an OCT image, predict/reconstruct its ground-truth structure on a pixel level. By solving this problem we would be able to interpret an OCT image completely and precisely without the help from a trained expert. It turns out that we can do much better. For simple structures we are able to reconstruct the ground-truth of an OCT image more than 98% correctly, and for more complicated structures (e.g., a multi-layered brain structure) we are looking at 93%. We achieved this through extensive uses of Machine Learning. The success of the Monte Carlo simulation already puts us in a great position by providing us with a great deal of data (effectively unlimited), in the form of (image, truth) pairs. Through a transformation of the high-dimensional response variable, we convert the learning task into a multi-output multi-class classification problem and a multi-output regression problem. We then build a hierarchy architecture of machine learning models (committee of experts) and train different parts of the architecture with specifically designed data sets. In prediction, an unseen OCT image first goes through a classification model to determine its structure (e.g., the number and the types of layers present in the image); then the image is handed to a regression model that is trained specifically for that particular structure to predict the length of the different layers and by doing so reconstruct the ground-truth of the image. We also demonstrate that ideas from Deep Learning can be useful to further improve the performance.

It is worth pointing out that solving the inverse problem automatically improves the imaging depth, since previously the lower half of an OCT image (i.e., greater depth) can be hardly seen but now becomes fully resolved. Interestingly, although OCT signals consisting the lower half of the image are weak, messy, and uninterpretable to human eyes, they still carry enough information which when fed into a well-trained machine learning model spits out precisely the true structure of the object being imaged. This is just another case where Artificial Intelligence (AI) outperforms human. To the best knowledge of the author, this thesis is not only a success but also the first attempt to reconstruct an OCT image at a pixel level. To even give a try on this kind of task, it would require fully annotated OCT images and a lot of them (hundreds or even thousands). This is clearly impossible without a powerful simulation tool like the one developed in this thesis.

Tsunamis - a model of their generation and propagation

A general solution is presented for water waves generated by an arbitrary movement of the bed (in space and time) in a two-dimensional fluid domain with a uniform depth. The integral solution which is developed is based on a linearized approximation to the complete (nonlinear) set of governing equations. The general solution is evaluated for the specific case of a uniform upthrust or downthrow of a block section of the bed; two time-displacement histories of the bed movement are considered.

An integral solution (based on a linear theory) is also developed for a three-dimensional fluid domain of uniform depth for a class of bed movements which are axially symmetric. The integral solution is evaluated for the specific case of a block upthrust or downthrow of a section of the bed, circular in planform, with a time-displacement history identical to one of the motions used in the two-dimensional model.

Since the linear solutions are developed from a linearized approximation of the complete nonlinear description of wave behavior, the applicability of these solutions is investigated. Two types of non-linear effects are found which limit the applicability of the linear theory: (1) large nonlinear effects which occur in the region of generation during the bed movement, and (2) the gradual growth of nonlinear effects during wave propagation.

A model of wave behavior, which includes, in an approximate manner, both linear and nonlinear effects is presented for computing wave profiles after the linear theory has become invalid due to the growth of nonlinearities during wave propagation.

An experimental program has been conducted to confirm both the linear model for the two-dimensional fluid domain and the strategy suggested for determining wave profiles during propagation after the linear theory becomes invalid. The effect of a more general time-displacement history of the moving bed than those employed in the theoretical models is also investigated experimentally.

The linear theory is found to accurately approximate the wave behavior in the region of generation whenever the total displacement of the bed is much less than the water depth. Curves are developed and confirmed by the experiments which predict gross features of the lead wave propagating from the region of generation once the values of certain nondimensional parameters (which characterize the generation process) are known. For example, the maximum amplitude of the lead wave propagating from the region of generation has been found to never exceed approximately one-half of the total bed displacement. The gross features of the tsunami resulting from the Alaskan earthquake of 27 March 1964 can be estimated from the results of this study.

Cytokine pathway disruption in a mouse model of schizophrenia induced by Munc18-1a overexpression in the brain

Background: An accumulating body of evidence points to the significance of neuroinflammation and immunogenetics in schizophrenia, and an imbalance of cytokines in the central nervous system (CNS) has been suggested to be associated with the disorder. Munc18-overexpressing mice (Munc18-OE) have provided a model for the study of the alterations that may underlie the symptoms of subjects with schizophrenia. The aim of the present study was to elucidate the involvement of neuroinflammation and cytokine imbalance in this model. Methods: Cytokines were evaluated in the cortex and the striatum of Munc18-OE and wild-type (WT) mice by enzyme-linked immunosorbent assay (ELISA). Protein levels of specific microglia and macrophage, astrocytic and neuroinflammation markers were quantified by western blot in the cortex and the striatum of Munc18-OE and WT mice. Results: Each cytokine evaluated (Interferon-gamma (IFN-gamma), Tumor Necrosis Factor-alpha (TNF-alpha), Interleukin-2 (IL-2) and CCL2 chemokine) was present at higher levels in the striatum of Munc18-OE mice than WT. Cortical TNF-alpha and IL-2 levels were significantly lower in Munc18-OE mice than WT mice. The microglia and macrophage marker CD11b was lower in the cortexes of Munc18-OE mice than WT, but no differences were observed in the striatum. Glial Fibrillary Acidic Protein (GFAP) and Nuclear Factor-kappaB (NF-kappa B)p65 levels were not different between the groups. Interleukin-1beta (IL-1 beta) and IL-6 levels were beneath detection limits. Conclusions: The disrupted levels of cytokines detected in the brain of Munc18-OE mice was found to be similar to clinical reports and endorses study of this type for analysis of this aspect of the disorder. The lower CD11b expression in the cortex but not in the striatum of the Munc18-OE mice may reflect differences in physiological activity. The cytokine expression pattern observed in Munc18-OE mice is similar to a previously published model of schizophrenia caused by maternal immune activation. Together, these data suggest a possible role for an immune imbalance in this disorder.

Brca1 is expressed in human microglia and is dysregulated in human and animal model of ALS

Background: There is growing evidence that microglia are key players in the pathological process of amyotrophic lateral sclerosis (ALS). It is suggested that microglia have a dual role in motoneurone degeneration through the release of both neuroprotective and neurotoxic factors. Results: To identify candidate genes that may be involved in ALS pathology we have analysed at early symptomatic age (P90), the molecular signature of microglia from the lumbar region of the spinal cord of hSOD1(G93A) mice, the most widely used animal model of ALS. We first identified unique hSOD1(G93A) microglia transcriptomic profile that, in addition to more classical processes such as chemotaxis and immune response, pointed toward the potential involvement of the tumour suppressor gene breast cancer susceptibility gene 1 (Brca1). Secondly, comparison with our previous data on hSOD1(G93A) motoneurone gene profile substantiated the putative contribution of Brca1 in ALS. Finally, we established that Brca1 protein is specifically expressed in human spinal microglia and is up-regulated in ALS patients. Conclusions: Overall, our data provide new insights into the pathogenic concept of a non-cell-autonomous disease and the involvement of microglia in ALS. Importantly, the identification of Brca1 as a novel microglial marker and as possible contributor in both human and animal model of ALS may represent a valid therapeutic target. Moreover, our data points toward novel research strategies such as investigating the role of oncogenic proteins in neurodegenerative diseases.

On the domain decomposition and transmission line modelling finite element method for time-domain induction motor analysis

This paper demonstrates how a finite element model which exploits domain decomposition is applied to the analysis of three-phase induction motors. It is shown that a significant gain in cpu time results when compared with standard finite element analysis. Aspects of the application of the method which are particular to induction motors are considered: the means of improving the convergence of the nonlinear finite element equations; the choice of symmetrical sub-domains; the modelling of relative movement; and the inclusion of periodic boundary conditions. © 1999 IEEE.

A three-dimensional tunnel model for calculation of train-induced ground vibration

The frequency range of interest for ground vibration from underground urban railways is approximately 20 to 100 Hz. For typical soils, the wavelengths of ground vibration in this frequency range are of the order of the spacing of train axles, the tunnel diameter and the distance from the tunnel to nearby building foundations. For accurate modelling, the interactions between these entities therefore have to be taken into account. This paper describes an analytical three-dimensional model for the dynamics of a deep underground railway tunnel of circular cross-section. The tunnel is conceptualised as an infinitely long, thin cylindrical shell surrounded by soil of infinite radial extent. The soil is modelled by means of the wave equations for an elastic continuum. The coupled problem is solved in the frequency domain by Fourier decomposition into ring modes circumferentially and a Fourier transform into the wavenumber domain longitudinally. Numerical results for the tunnel and soil responses due to a normal point load applied to the tunnel invert are presented. The tunnel model is suitable for use in combination with track models to calculate the ground vibration due to excitation by running trains and to evaluate different track configurations. © 2006 Elsevier Ltd. All rights reserved.

Hafnia nanoparticles - a model system for graphene growth on a dielectric

We study graphene growth on hafnia (HfO2) nanoparticles by chemical vapour deposition using optical microscopy, high resolution transmission electron microscopy and Raman spectroscopy. We find that monoclinic HfO2 nanoparticles neither reduce to a metal nor form a carbide while nucleating nanometer domain-sized few layer graphene. Hence we regard this as an interesting non-metallic catalyst model system with the potential to explore graphene growth directly on a (high-k) dielectric. HfO2 nanoparticles coated with few layer graphene by atmospheric pressure CVD with methane and hydrogen at 950 °C. (© 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) Graphene growth on hafnia (HfO2) nanoparticles by chemical vapour deposition (CVD) is studied. It is found that monoclinic HfO2 nanoparticles neither reduce to a metal nor form a carbide while nucleating nanometer domain-sized few layer graphene. Hence the authors of this Letter regard this as an interesting non-metallic catalyst model system with the potential to explore graphene growth directly on a (high-k) dielectric. © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Angular dependence of domain wall resistivity in artificial magnetic domain structures.

We exploit the ability to precisely control the magnetic domain structure of perpendicularly magnetized Pt/Co/Pt trilayers to fabricate artificial domain wall arrays and study their transport properties. The scaling behavior of this model system confirms the intrinsic domain wall origin of the magnetoresistance, and systematic studies using domains patterned at various angles to the current flow are excellently described by an angular-dependent resistivity tensor containing perpendicular and parallel domain wall resistivities. We find that the latter are fully consistent with Levy-Zhang theory, which allows us to estimate the ratio of minority to majority spin carrier resistivities, rho downward arrow/rho upward arrow approximately 5.5, in good agreement with thin film band structure calculations.

Vowel normalisation: Time-domain processing of the internal dynamics of speech

Human listeners can identify vowels regardless of speaker size, although the sound waves for an adult and a child speaking the ’same’ vowel would differ enormously. The differences are mainly due to the differences in vocal tract length (VTL) and glottal pulse rate (GPR) which are both related to body size. Automatic speech recognition machines are notoriously bad at understanding children if they have been trained on the speech of an adult. In this paper, we propose that the auditory system adapts its analysis of speech sounds, dynamically and automatically to the GPR and VTL of the speaker on a syllable-to-syllable basis. We illustrate how this rapid adaptation might be performed with the aid of a computational version of the auditory image model, and we propose that an auditory preprocessor of this form would improve the robustness of speech recognisers.

Measurement of the Interaction Between Recombinant I-domain from Integrin alpha 2 beta 1 and a Triple Helical Collagen Peptide with the GFOGER Binding Motif Using Molecular Force Spectroscopy.

The role of the collagen-platelet interaction is of crucial importance to the haemostatic response during both injury and pathogenesis of the blood vessel wall. Of particular interest is the high affinity interaction of the platelet transmembrane receptor, alpha 2 beta 1, responsible for firm attachment of platelets to collagen at and around injury sites. We employ single molecule force spectroscopy (SMFS) using the atomic force microscope (AFM) to study the interaction of the I-domain from integrin alpha 2 beta 1 with a synthetic collagen related triple-helical peptide containing the high-affinity integrin-binding GFOGER motif, and a control peptide lacking this sequence, referred to as GPP. By utilising synthetic peptides in this manner we are able to study at the molecular level subtleties that would otherwise be lost when considering cell-to-collagen matrix interactions using ensemble techniques. We demonstrate for the first time the complexity of this interaction as illustrated by the complex multi-peaked force spectra and confirm specificity using control blocking experiments. In addition we observe specific interaction of the GPP peptide sequence with the I-domain. We propose a model to explain these observations.

An efficient model for the dynamic behaviour of a single pile in viscoelastic soil

This paper presents a novel, three-dimensional, single-pile model, formulated in the wavenumber domain and adapted to account for boundary conditions using the superposition of loading cases. The pile is modelled as a column in axial vibration, and a Euler-Bernoulli beam in lateral vibration. The surrounding soil is treated as a viscoelastic continuum. The response of the pile is presented in terms of the stiffness and damping coefficients, and also the magnitude and phase of the pile-head frequency-response function. Comparison with existing models shows that excellent agreement is observed between this model, a boundary-element formulation, and an elastic-continuum-type formulation. This three-dimensional model has an accuracy equivalent to a 3D boundary-element model, and a runtime similar to a 2D plane-strain analytical model. Analysis of the response of the single pile illustrates a difference in axial and lateral vibration behaviour; the displacement along the pile is relatively invariant under axial loads, but in lateral vibration the pile exhibits localised deformations. This implies that a plane-strain assumption is valid for axial loadings and only at higher frequencies for lateral loadings. © 2013 Elsevier Ltd.

Antiwindup design for induction motor control in the field weakening domain

Operation of induction machines in the high-speed and/or high-torque range requires field-weakening to comply with voltage and current physical limitations. This paper presents an anti-windup approach to this problem: rather than developing an ad-hoc field weakening strategy in the high-speed region, we equip an unconstrained vector-control design with an anti-windup module that automatically adjusts the current and flux set-points so that voltage and current constraints are satisfied at every operating point. The anti-windup module includes a feedforward modification of the set point aimed at maximizing the available torque in steady-state and a feedback modification of the controller based on an internal model-based antiwindup scheme. This paper includes a complete stability analysis of the proposed solution and presents encouraging experimental results on an industrial drive. © 2012 IEEE.

An Accurate and Fast Behavioral Model for PLL Frequency Synthesizer Phase Noise/Spurs Prediction

This paper presents a behavior model for PLL Frequency Synthesizer. All the noise sources are modeled with noise voltages or currents in time-domain. An accurate VCO noise model is introduced, including both thermal noise and 1/f noise. The behavioral model can be co-simulated with transistor level circuits with fast speed and provides more accurate phase noise and spurs prediction. Comparison shows that simulation results match very well with measurement results.