The hippocampus is a complex brain structure that has been studied extensively and is subject to abnormal structural change in various neuropsychiatric disorders. The highest definition in vivo method of visualizing the anatomy of this structure is structural Magnetic Resonance Imaging (MRI). Gross structure can be assessed by the naked eye inspection of MRI scans but measurement is required to compare scans from individuals within normal ranges, and to assess change over time in individuals. The gold standard of such measurement is manual tracing of the boundaries of the hippocampus on scans. This is known as a Region Of Interest (ROI) approach. ROI is laborious and there are difficulties with test-retest and inter-rater reliability. These difficulties are primarily due to uncertainty in designation of the hippocampus boundary. An improved, less labour intensive and more reliable method is clearly desirable. This thesis describes a fully automated hybrid methodology that is able to first locate and then extract hippocampal volumes from 3D 1.5T MRI T1 brain scans automatically. The hybrid algorithm uses brain atlas mappings and fuzzy inference to locate hippocampal areas and create initial hippocampal boundaries. This initial location is used to seed a deformable manifold algorithm. Rule based deformations are then applied to refine the estimate of the hippocampus locations. Finally, the hippocampus boundaries are corrected through an inference process that assures adherence to an expected hippocampus volume. The ICC values of this methodology when compared to the manual segmentation of the same hippocampi result in a 0.73 for the left and 0.81 for the right hippocampi. These values both fall within the range of reliability testing according to the manual ‘gold standard’ technique. Thus, this thesis describes the development and validation of a genuinely automated approach to hippocampal volume extraction of potential utility in studies of a range of neuropsychiatric disorders and could eventually find clinical applications.

Skin cancer is the most frequent malignant neoplasm for Caucasian individuals. According to the Skin Cancer Foundation, the incidence of melanoma, the most malignant of skin tumours, and resultant mortality, have increased exponentially during the past 30 years, and continues to grow. [1]. Although often intractable in advanced stages, skin cancer in general and melanoma in particular, if detected in an early stage, can achieve cure ratios of over 95% [1,55]. Early screening of the lesions is, therefore, crucial, if a cure is to be achieved. Most skin lesions classification systems rely on a human expert supported dermatoscopy, which is an enhanced and zoomed photograph of the lesion zone. Nevertheless and although contrary claims exist, as far as is known by the author, classification results are currently rather inaccurate and need to be verified through a laboratory analysis of a piece of the lesion’s tissue. The aim of this research was to design and implement a system that was able to automatically classify skin spots as inoffensive or dangerous, with a small margin of error; if possible, with higher accuracy than the results achieved normally by a human expert and certainly better than any existing automatic system. The system described in this thesis meets these criteria. It is able to capture an unconstrained image of the affected skin area and extract a set of relevant features that may lead to, and be representative of, the four main classification characteristics of skin lesions: Asymmetry; Border; Colour; and Diameter. These relevant features are then evaluated either through a Bayesian statistical process - both a simple k-Nearest Neighbour as well as a Fuzzy k-Nearest Neighbour classifier - a Support Vector Machine and an Artificial Neural Network in order to classify the skin spot as either being a Melanoma or not. The characteristics selected and used through all this work are, to the author’s knowledge, combined in an innovative manner. Rather than simply selecting absolute values from the images characteristics, those numbers were combined into ratios, providing a much greater independence from environment conditions during the process of image capture. Along this work, image gathering became one of the most challenging activities. In fact several of the initially potential sources failed and so, the author had to use all the pictures he could find, namely on the Internet. This limited the test set to 136 images, only. Nevertheless, the process results were excellent. The algorithms developed were implemented into a fully working system which was extensively tested. It gives a correct classification of between 76% and 92% – depending on the percentage of pictures used to train the system. In particular, the system gave no false negatives. This is crucial, since a system which gave false negatives may deter a patient from seeking further treatment with a disastrous outcome. These results are achieved by detecting precise edges for every lesion image, extracting features considered relevant according to the giving different weights to the various extracted features and submitting these values to six classification algorithms – k-Nearest Neighbour, Fuzzy k-Nearest Neighbour, Naïve Bayes, Tree Augmented Naïve Bayes, Support Vector Machine and Multilayer Perceptron - in order to determine the most reliable combined process. Training was carried out in a supervised way – all the lesions were previously classified by an expert on the field before being subject to the scrutiny of the system. The author is convinced that the work presented on this PhD thesis is a valid contribution to the field of skin cancer diagnostics. Albeit its scope is limited – one lesion per image – the results achieved by this arrangement of segmentation, feature extraction and classification algorithms showed this is the right path to achieving a reliable early screening system. If and when, to all these data, values for age, gender and evolution might be used as classification features, the results will, no doubt, become even more accurate, allowing for an improvement in the survival rates of skin cancer patients.

An investigation into the effects of deforestation and habitat fragmentation on a central Amazonian frog community was carried out on the Biological Dynamics of Forest Fragments Project, (BDFFP) based in Manaus, Amazonas, Brazil. Three aspects of deforestation were investigated: the effects of fragmentation and reserve size, the influence of matrix habitat and the effects offorest edge on frogs in 10- hectare isolates. Very few frogs species were lost from fragments (regardless of size) following fragmentation and isolation for 7-10 years. Instead, frog species richness increased, with an average increase in 10 species per fragment. The increase in species was brought about by an increase in matrix associated species, some of which were present in the absence of their preferred breeding habitat. Different frog species had varied responses to fragmentation. The abundance of Eleutherodactylus fenestratus, a terrestrial breeder, increased significantly as fragment size decreased, and its abundance was significantly higher in both large and small fragments than in continuous forest. Colostethus stepheni, a semi-terrestrial breeder, was less abundant in fragments than continuous forest. Finally, Eleutherodactylus zimmermanae and Osteocephalus sp.A did not differ significantly in abundance among fragments, or between fragments and continuous forest. Multiple regression analysis indicated that variation in litter depth and canopy cover may explain the observed increase in E. fenestratus abundance in small fragments. Breeding success of pool breeders attracted to artificial pools was variable, but there was no evidence of reduced breeding success in fragments relative to primary forest. Overall, fragmentation appeared to affect the frog community less severely than other taxonomic groups. Species richness in tropical forest remnants showed a tendency to increase as a result of fragmentation, and only one of four species exhibited lowered abundance in fragments. The effects of fragmentation and subsequent small isolated populations of frog species in fragments was alleviated somewhat by the ability of central Amazonian frogs to utilise matrix habitat. Litter frogs, wallow breeders and upland forest pool breeders all displayed an ability to permeate, and in most cases become residents of matrix habitat. Both matrix with a history of cut and burn and matrix which was only cut supported a similar frog community. Thus, for central Amazonian frogs isolates were not truly isolated due to the ability of frogs to utilise matrix habitat. Litter frogs in 10- hectare isolates displayed an independence from edge related abiotic and biotic gradients. This independence was true for frog species richness, abundance and composition. Consequently, forest fragments are in effect larger for frogs than for other taxa who are constrained by edge effects to exist only within the core of large fragments. Colostethus stepheni was one species found to be negatively affected by habitat fragmentation. Abundance of Colostethus stepheni was significantly lower in fragments and matrix habitat compared to primary continuous forest. This species was also found to be sensitive to edge, with higher abundances recorded as distance from the edge increased. The community level approach adopted in this study, as opposed to intensive investigation of single species, may well have overlooked other' sensitive' species which show more subtle responses to habitat fragmentation than that of Colostethus stepheni. The ability to use matrix and the relative independence from edge related phenomena accounts for the resilience of central Amazonian frogs in a disturbed and fragmented landscape. However, with respect to matrix habitat, forest surveyed in this study has not been repeatedly cut or cut and burnt (with the exception of pasture-land where the frog community is depauperate). In other areas of the Amazon, away from the experimental plots of the BDFFP, matrix habitat is destroyed more frequently. More work is needed to determine the ability of central Amazonian frogs to utilise matrix habitat which is disturbed more frequently.

Accurate, spatially resolved wind measurements in the lower atmosphere are critical to improving current weather forecasting models. Wind shear detection for midsized airports, not covered under the Federal Aviation Administration's (FAA's) Terminal Doppler Weather Radars, would significantly reduce personal aircraft accidents. Atmospheric dynamics studies would also benefit from high accuracy, spatially resolved wind profiles within the planetary boundary layer.This dissertation discusses a thorough investigation of a novel alternating frequency incoherent Doppler lidar method named, the Alternating Edge Technique. After discussing the necessary background, a theoretical development of how the Alternating Edge Technique can be used to estimate the molecular to aerosol backscatter ratio is presented. The ability to estimate the molecular component vastly improves the accuracy of wind measurements, and adds additional information about the atmosphere being probed. A detailed modeling program was developed to evaluate the expected performance of this instrument, and to allow comparisons to be made of various components and operating conditions. Several of the key components are then evaluated experimentally, and these results are used to perform realistic Monte Carlo simulations in order to evaluate the ability of the Alternating edge technique, using available components, to estimate the molecular component of the backscattered light, and to accurately estimate average wind speed.

This thesis will study a variety of problems in graph theory. Initially, the focus will be on finding minimal degree conditions which guarantee the existence of various subgraphs. These subgraphs will all be formed of cycles, and this area of work will fall broadly into two main categories. First to be considered are cycles in edge-coloured graphs and, in particular, two questions of Li, Nikiforov and Schelp. It will be shown that a 2-edge-coloured graph with minimal degree at least 3n/4 either is isomorphic to the complete 4-partite graph with classes of order n/4, or contains monochromatic cycles of all lengths between 4 and n/2 (rounded up). This answers a conjecture of Li, Nikiforov and Schelp. Attention will then turn to the length of the longest monochromatic cycle in a 2-edge-coloured graph with minimal degree at least cn. In particular, a lower bound for this quantity will be proved which is asymptotically best possible. The next chapter of the thesis then shows that a hamiltonian graph with minimal degree at least (5-sqrt7)n/6 contains a 2-factor with two components. The thesis then concludes with a chapter about X_H, which is the number of copies of a graph H in the random graph G(n,p). In particular, it will be shown that, for a connected graph H, the value of X_H modulo k is approximately uniformly distributed, provided that k is not too large a function of n.

In their recent paper ``Edge-transitive products," Hammack, Imrich, and Klavzar showed that the direct product of connected, non-bipartite graphs is edge-transitive if and only if both factors are edge-transitive, and at least one is arc-transitive. However, little is known when the product is bipartite. This thesis extends this result (in part) for the case of bipartite graphs using a new technique called "stacking." For R-thin, connected, bipartite graphs A and B, we show that A x B is arc-transitive if and only if A and B are both arc-transitive. Further, we show A x B is edge-transitive only if at least one of A, B is also edge-transitive, and give evidence that strongly suggests that in fact both factors must be edge-transitive.

The Biloxi Marsh of the eastern Mississippi River delta plain is exposed to meteorological forces ranging from large-scale, mid-latitude cyclones to smaller scale storms and squalls. Each time that these marsh platforms are exposed to a storm event, the potential exists for either deposition or erosion to take place. This study examines the connection between wind speeds, stratigraphic composition, marsh edge morphology, and marsh edge erosion at 4 sites in the Biloxi Marsh. As much as 2.17 m of erosion were measured during the 9-month study with a maximum, averaged erosion rate of 0.03 m day-1. Shell berm transgression was also documented as the result of winds from a low-pressure system near the site that resulted large waves. Shell berm movement occurred as a result of high wind speeds on shore but there was no connection between wind speeds and erosion for the duration of this study across all sites.

SNA provides a wide range of tools that allow examination of telecommunications graphs. Those graphs contain vertices representing cell phone users and lines standing for established connections. Many sna tools do not incorporate the intensity of interaction. This may lead to wrong conclusions because the difference between best friends and random contacts can be defined by the accumulated duration of talks. To solve this problem, we propose a closeness centrality measure (ewc) that incorporates line values and compare it to Freeman's closeness. Small exemplary networks will demonstrate the characteristics of the weighted closeness compared to other centrality measures. Finally, the ewc will be tested on a real-world telecommunications graph provided by a large Austrian mobile service provider and the advantages of the ewc will be discussed. / Series: Research Report Series / Department of Statistics and Mathematics

This report deals with reconstruction of strain fields within deformed materials. The method relies upon data generated from Bragg edge measurements, in which information is gained from neutron beams that are sent through the sample. The reconstruction has been made by modelling the strain field as a Gaussian process, assigned a covariance structure customized by incorporation of the so-called equilibrium constraints. By making use of an approximation scheme well suited for the problem, the complexity of the computations has been significantly reduced. The results from numerical simulations indicates a better performance as compared to previous work in this area.

Thesis advisor: Michael J. Naughton / The nanoscale morphology of metals can enable special functionality in plasmonic and electrochemical devices, with applications in energy conversion and storage, sensors, and computers. In particular, sharp edges on metal nano and microstructures are understood to affect the density of electrons on the metal surface. The associated concentration of electric field can concentrate surface plasmon polaritons (SPPs) and enable waveguiding of the SPPs, as we show in this thesis for sharp ridges along aluminum nanowires. Also important is the presence of facets on the metal structures, which determines the orbitals that electrons occupy on the metal surface. Changes in both the electron density and orbitals can affect the binding of molecules to the metal, which can improve reaction kinetics in catalysis. We demonstrate this on gold dendrite and plate electrocatalysts for CO2 electrolysis. Regarding metal nanostructure fabrication, electrochemical deposition and corrosion have demonstrated promising control over the morphology, including the topography, crystallinity, grain boundaries, and crystal faceting. This is important, because existing methods for metal nanostructure fabrication can only produce a circumscribed assortment of morphologies. In contrast, semiconductors and insulators have many new deposition techniques that produce a wide range of controlled morphologies. Of further appeal, electrochemical techniques are solution-based and typically operate at room temperature and pressure, allowing facile scale-up to industrial production. Here we demonstrate and discuss the mechanisms of two new techniques, which produce the aluminum nanowires and gold dendrites and plates discussed above. / Thesis (PhD) — Boston College, 2018. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Physics.