This dissertation serves as a unifying document for three related articles developed during my dissertation research. The projects involve the development of single-subject transcriptome (i.e. gene expression data) methodology for precision medicine and related applications. Traditional statistical approaches are largely unavailable in this setting due to prohibitive sample size and lack of independent replication. This leads one to rely on informatic devices including knowledgebase integration (e.g., gene set annotations) and external data sources (e.g., estimation of inter-gene correlation). Common statistical themes include multivariate statistics (such as Mahalanobis distance and copulas) and large-scale significance testing. Briefly, the first work describes the development of clinically relevant single-subject metrics of gene set (pathway) differential expression, N-of-1-pathways Mahalanobis distance (MD) scores. Next, the second article describes a method which overcomes a major shortcoming of the MD framework by accounting for inter-gene correlation. Lastly, the statistics developed in the previous works are re-purposed to analyze single-cell RNA-sequencing data derived from rare cells. Importantly, these works represent an interdisciplinary effort and show that creative solutions for pressing issues become possible at the intersection of statistics, biology, medicine, and computer science.

Volatility plays an important role in option pricing and risk management. It is crucial that volatility is modelled as accurately as possible in order to forecast with confidence. The challenge is in the selection of the ‘best’ model with so many available models and selection criteria. The Model Confidence Set (MCS) solves this problem by choosing a group of models that are equally good. A set of GARCH models were estimated for several JSE indices and the MCS was used to trim the group of models to a subset of equally superior models. Using the Mean Squared Error to evaluate the relative performance of the MCS, GARCH (1,1) and Random Walk, it was found that the MCS, with an equally weighted combination of models, performed better than the GARCH (1,1) and Random Walk for instances where volatility in the returns data was high. For instances of low volatility in the returns, the GARCH (1,1) had superior 5-day forecasts but the MCS had better performance for 10-days and greater. The EGARCH (2,1) volatility model was selected by the MCS for 5 out of the 6 indices as the most superior model. The Random Walk was shown to have better long term forecasting performance.

We identify three structures that lie in the parameter plane of the rational map F(z) = zⁿ + λ / zᵈ, for which z is a complex number, λ a complex parameter, n ≥ 4 is even, and d ≥ 3 is odd. There exists a Sierpindelbrot arc, an infinite sequence of pairs of Mandelbrot sets and Sierpinski holes, that limits to the parameter at the end of the arc. There exists as well a qualitatively different Sierpindelbrot arc, an infinite sequence of pairs of Mandelbrot sets and Sierpinski holes, that limits to the parameter at the center of the arc. Furthermore, there exist infinitely many arcs of each type. A parameter can travel along a continuous path from the Cantor set locus, along infinitely many arcs of the first type in a successively smaller region of the parameter plane, while passing through an arc of the second type, to the parameter at the center of the latter arc. This infinite sequence of Sierpindelbrot arcs is a Sierpinski Mandelbrot spiral.

by He Xinggang. / "December 2000." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2000. / Includes bibliographical references (p. 69-73). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

This dissertation presents results for two separate problems, both in the context of variational fracture models. The first problem involved developing and analyzing models of fracture in which we modeled the energy dissipated by crack growth as concentrated on the front of the crack. While many engineering models of fracture are based on a notion of crack front, there had not been a rigorous definition. We present the first work in this area, which includes a natural weak definition of crack front and front speed, a model of fracture whose evolution is described at the crack front, and a relaxation result that shows that these front based dissipations are all effectively equivalent to a Griffith-type dissipation. The second problem involved the computation of stationary points for Mumford-Shah and fracture using a level set method. Our method improves on existing techniques in that it can handle tips in the singular set and can find minimizers that previous techniques are unable to resolve.

Our method can be naturally applied in the application of manga stylization rendering. We propose a novel colorization technique that propagates color over manga regions exhibiting pattern-continuity as well as intensity-continuity. The proposed method works effectively on colorizing black-and-white manga which contains intensive amount of strokes, hatching, halftoning and screening. Once the user scribbles on the drawing, a local, statistical based pattern feature obtained with Gabor wavelet filters is applied to measure the pattern-continuity. The boundary is then propagated by the level set method that monitors the pattern-continuity. Regions with open boundaries or multiple disjointed regions with similar patterns can be sensibly segmented by a single scribble. With the segmented regions, various colorization techniques can be applied to replace colors, colorize with stroke preservation, or even convert pattern to shading. Based on the observation of the manga features, we then propose a framework to generate manga-style backgrounds from real photographs. It frees manga artists from the tedious and time-consuming background production. To mimic how manga artists produce the tidy background, it composes of two major steps, the line drawing and the screen laying. A line importance model is proposed to simplify and embolden lines in a stylish way. During the screen laying, texture analysis is utilized to automatically match the regions in photographs with screens in the database. Two types of screening mechanisms are proposed for matching shading & texture as well as the high-level structures. / Segmentation has always been a crucial and challenging topic in various computer vision and graphics applications. Information from image features can be incorporated into image segmentation techniques to make the process more accurate and robust. This dissertation focuses on encoding these feature information into the image processing procedures, including image segmentation and manga stylization rendering. / The fundamental part of this dissertation includes the discussion on the construction of the speed function, which is important in applying the curve-evolution based image segmentation. We firstly investigate the curvature term in the speed function, and then show how to transform the image segmentation problem into an interface propagating problem. We propose two formulations to enhance the speed function in level set methods, in order to tackle the segmentation problem of tagged MR images. First, a relaxation factor is introduced, aiming at relaxing the boundary condition when the boundary is unclear or blurry. Second, in order to incorporate human visual sensitive information from the image, a simple and general model is introduced to incorporate shape, texture and color features. By further extending this model, we present a unified approach for segmenting and tracking of the high-resolution color anatomical Chinese Visible Human (CVH) data. The underlying relationship of these two applications relies on the proposed variational framework for the speed function. Our proposed method can be used to segment the first slice of the volume data. Then based on the extracted boundary on the first slice, our method can also be adapted to track the boundary of the homogeneous organs among the subsequent serial images. In addition to the promising segmentation results, the tracking procedure requires only a small amount of user intervention. / Qu Yingge. / "July 2007." / Advisers: Heng Pheng Ann; Wong Tien Tsin. / Source: Dissertation Abstracts International, Volume: 69-01, Section: B, page: 0433. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 107-120). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.

Zheng Xi Yin. / "December 2002." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (p. 117-121). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

by Li Jian Lin. / "November 2004." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (p. 85-90) / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Lai, Chun Kit. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 83-87). / Abstract also in Chinese. / Chapter 1 --- Introduction --- p.6 / Chapter 2 --- Spectral sets in Rd --- p.11 / Chapter 2.1 --- Preliminaries --- p.11 / Chapter 2.2 --- Fundamental domains and convex sets in Rd --- p.15 / Chapter 2.3 --- Finite union of cubes --- p.20 / Chapter 3 --- Spectral theory on discrete groups --- p.27 / Chapter 3.1 --- Finite groups and Zd --- p.28 / Chapter 3.2 --- Rational spectrums and tiling sets --- p.32 / Chapter 3.3 --- Fuglede´ةs Problem in R1 --- p.37 / Chapter 3.4 --- "Failure of Fuglede´ةs Conjecture in Rd, d >3" --- p.42 / Chapter 4 --- Self-similar tiles in R1 --- p.49 / Chapter 4.1 --- Basics of self-similar tiles --- p.49 / Chapter 4.2 --- Self-similar tile digit sets and spectral problem --- p.52 / Chapter 4.3 --- Kenyon criterion --- p.55 / Chapter 5 --- Spectral self-similar measures --- p.66 / Chapter 5.1 --- Spectral self-similar measures --- p.66 / Chapter 5.2 --- One-dimensional self-similar measures --- p.72 / Chapter 5.3 --- General properties of spectral measures --- p.80 / Bibliography --- p.83

Wong, Wing Chun. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 66-67). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Background Study --- p.4 / Chapter 2.1 --- Basic Definitions and Notations --- p.5 / Chapter 2.2 --- Tarjan and Trojanowski's algorithm --- p.6 / Chapter 2.2.1 --- Techniques --- p.6 / Chapter 2.2.2 --- Algorithm --- p.8 / Chapter 2.3 --- "Fomin, Grandoni and Kratsch's Algorithm" --- p.9 / Chapter 2.3.1 --- Techniques --- p.9 / Chapter 2.3.2 --- Algorithm --- p.14 / Chapter 3 --- Improvements --- p.18 / Chapter 3.1 --- Tarjan and Trojanowski´ةs Algorithm --- p.18 / Chapter 3.1.1 --- Correctness and Running Time Analysis --- p.28 / Chapter 3.1.2 --- Improvement --- p.30 / Chapter 3.1.3 --- Using more weights --- p.35 / Chapter 3.2 --- The First Algorithm --- p.37 / Chapter 3.2.1 --- Standard Analysis --- p.37 / Chapter 3.2.2 --- Measure and Conquer --- p.38 / Chapter 3.2.3 --- Using more weights --- p.42 / Chapter 3.3 --- The Second Algorithm --- p.43 / Chapter 3.3.1 --- Running Time Analysis --- p.44 / Chapter 3.3.2 --- Using More Weights --- p.45 / Chapter 3.4 --- The Third Algorithm --- p.46 / Chapter 4 --- Lower Bounds --- p.52 / Chapter 4.1 --- Tarjan and Trojanowski's Algorithm --- p.52 / Chapter 4.2 --- The First Algorithm --- p.55 / Chapter 4.3 --- The Second Algorithm --- p.58 / Chapter 4.4 --- The Third Algorithm --- p.60 / Chapter 5 --- Conclusion --- p.63 / Bibliography --- p.66