This thesis studies survival analysis techniques dealing with censoring to produce predictive tools that predict the risk of endovascular aortic aneurysm repair (EVAR) re-intervention. Censoring indicates that some patients do not continue follow up, so their outcome class is unknown. Methods dealing with censoring have drawbacks and cannot handle the high censoring of the two EVAR datasets collected. Therefore, this thesis presents a new solution to high censoring by modifying an approach that was incapable of differentiating between risks groups of aortic complications. Feature selection (FS) becomes complicated with censoring. Most survival FS methods depends on Cox's model, however machine learning classifiers (MLC) are preferred. Few methods adopted MLC to perform survival FS, but they cannot be used with high censoring. This thesis proposes two FS methods which use MLC to evaluate features. The two FS methods use the new solution to deal with censoring. They combine factor analysis with greedy stepwise FS search which allows eliminated features to enter the FS process. The first FS method searches for the best neural networks' configuration and subset of features. The second approach combines support vector machines, neural networks, and K nearest neighbor classifiers using simple and weighted majority voting to construct a multiple classifier system (MCS) for improving the performance of individual classifiers. It presents a new hybrid FS process by using MCS as a wrapper method and merging it with the iterated feature ranking filter method to further reduce the features. The proposed techniques outperformed FS methods based on Cox's model such as; Akaike and Bayesian information criteria, and least absolute shrinkage and selector operator in the log-rank test's p-values, sensitivity, and concordance. This proves that the proposed techniques are more powerful in correctly predicting the risk of re-intervention. Consequently, they enable doctors to set patients’ appropriate future observation plan.

Understanding how the brain works remains one of the key challenges for scientists. To further this understanding a wide variety of technologies and research methods have been developed. One such technology is conductive electrodes, used to measure the electrical signals elicited from neuronal cells and tissues. These electrodes can be fabricated as a singular electrode or as a multi-electrode array (MEA). This permits bio-electrical measurements from one particular area or simultaneous measurements from multiple areas, respectively. Studying electrical and chemical signals of individual cells in situ requires the use of electrodes with ≤20 µm diameter. However, electrodes of this size generally produce high impedance, perturbing recording of the small signals generated from individual cells. Nanomaterials, such as carbon nanotubes (CNTs), can be deposited to increase the real surface area of these electrodes, producing higher sensitivity measurements. This thesis investigates the potential for using photo-thermal chemical vapour deposition grown CNTs as the electrode material for a de novo fabricated MEA. This device aimed to measure electrochemical signals in the form of dopamine, an important mammalian neurotransmitter, as well as conventional bio-electrical signals that the device is designed for. Realising this aim began with improving CNT aqueous wetting behaviour via oxygen plasma functionalisation. This procedure demonstrated grafting of oxygen functional groups to the CNT structure, and dramatic improvements in aqueous wetting behaviour, with CNTs attached to the device. Subsequently, oxygen plasma functionalised CNT-based MEAs were fabricated and tested, allowing comparisons with a non-functionalised CNT MEA and a state-of-the-art commercial MEA. The functionalised CNT MEA demonstrated an order of magnitude improvement compared to commercial MEAs (2.75 kΩ vs. 25.6 kΩ), at the biologically relevant frequency of 1 kHz. This was followed by measurement of one of the best sensitivity density values, compared to the available literature, for the electrochemical detection of dopamine (9.48 µA µM-1 mm-2). The functionalised CNT MEA then illustrated some selectivity compared to common interferents, i.e. ascorbic acid, of a higher concentration. Nonetheless, imaging of the MEA revealed CNTs were being removed from the electrode areas due to extensive use. Therefore, the final results chapter aimed to develop a novel fabrication route for CNT-based MEAs that produced improved CNT retention on the electrodes. This next-generation functionalised CNT-based MEA displayed improved CNT retention, whilst also producing competitive electrochemical impedance values at 1 kHz (17.8 kΩ) and excellent electrochemical selectivity for dopamine vs. ascorbic acid. Overall, this thesis demonstrates the potential for using MEAs as electrochemical detectors of biological molecules, specifically when using functionalised CNTs as the electrode material.

One of the earlier works on eigen-based techniques for the hyper-complex domain of quaternions was on “quaternion principal component analysis of colour images”. The results of this work are still instructive in many aspects. First, it showed how naturally the quaternion domain accounts for the coupling between the dimensions of red, blue and green of an image, hence its suitability for multichannel processing. Second, it was clear that there was a lack of eigen-based techniques for such a domain, which explains the non-trivial gap in the literature. Third, the lack of such eigen-based quaternion tools meant that the scope and the applications of quaternion signal processing were quite limited, especially in the field of biomedicine. And fourth, quaternion principal component analysis made use of complex matrix algebra, which reminds us that the complex domain lays the building blocks of the quaternion domain, and therefore any research endeavour in quaternion signal processing should start with the complex domain. As such, the first contribution of this thesis lies in the proposition of complex singular spectrum analysis. That research provided a deep understanding and an appreciation of the intricacies of the complex domain and its impact on the quaternion domain. As the complex domain offers one degree of freedom over the real domain, the statistics of a complex variable x has to be augmented with its complex conjugate x*, which led to the term augmented statistics. This recent advancement in complex statistics was exploited in the proposed complex singular spectrum analysis. The same statistical notion was used in proposing novel quaternion eigen-based techniques such as the quaternion singular spectrum analysis, the quaternion uncorrelating transform, and the quaternion common spatial patterns. The latter two methods highlighted an important gap in the literature – there were no algebraic methods that solved the simultaneous diagonalisation of quaternion matrices. To address this issue, this thesis also presents new fundamental results on quaternion matrix factorisations and explores the depth of quaternion algebra. To demonstrate the efficacy of these methods, real-world problems mainly in biomedical engineering were considered. First, the proposed complex singular spectrum analysis successfully addressed an examination of schizophrenic data through the estimation of the event-related potential of P300. Second, the automated detection of the different stages of sleep was made possible using the proposed quaternion singular spectrum analysis. Third, the proposed quaternion common spatial patterns facilitated the discrimination of Parkinsonian patients from healthy subjects. To illustrate the breadth of the proposed eigen-based techniques, other areas of applications were also presented, such as in wind and financial forecasting, and Alamouti-based communication problems. Finally, a preliminary work is made available to suggest that the next step from this thesis is to move from static models (eigen-based models) to dynamic models (such as tracking models).

Bioresorbable polymers have been successfully used in clinical applications for many decades. They are those types of polymers that degrade into the human body and often don’t need to be removed out of the body, because their degradation products metabolise and enter the general metabolic pathways. However, there has been an increasing demand for better reliability and degradation control of bioresorbable polymeric devices causing researchers to abandon trial-and-error approaches to model-based methods. The mathematical or computer-based techniques for modelling of the mechanical properties are currently in their infancy or non-existent. This study aims to build a model to express the change in mechanical properties and detecting the degradation distribution within degrading bioresorbable polymers. It consists of three main parts. The first part reviews the literature for the most commonly used bioresorbable polymers and their applications. It also reviews the existing mathematical models for biodegradation. The experimental data of six PLLA films are also reviewed to provide insight into changes in mechanical properties of degrading bioresorbable polymers during hydrolytic degradation. The review shows that the mechanical properties are highly affected by the changes in molecular weight and crystallization. The second part presents a constitutive law for prediction of the elastic moduli, tensile strength and Poisson’s ratio of amorphous and semi-crystalline bioresorbable polymers based on the novel idea of formation cavity and crystal inclusions within degrading bioresorbable polymers. The results of using the constitutive law show that it can fit the experimental data fairly well. The third part presents a vibration-based study that shows the curvature mode shapes can successfully reveal the degradation distribution within, for instance, a simple cantilever beam or a coronary stent. This study also presents a chapter for computer modelling of the degradation behaviour of polyester-based tissue scaffolds using a degradation model developed in the University of Leicester.

No description available.

No description available.

The aim of this research was to develop an array of conducting polymer gas sensors as part of an electronic nose designed for monitoring the metabolites produced from the bacteria present in wounds. The device was designed to be a portable system that could discriminate between relevant bacteria non-invasively using solid phase microextraction and an array of conducting polymers and metal oxide sensors in conjunction with pattern-recognition software. In order to develop the sensors, GC/MS headspace analysis of a selection of bacterial species that are most commonly found to be present in wounds was first undertaken in order to determine the volatile key markers. The chosen key markers were then used as calibration gases in order to test and develop the sensors. Electrochemical techniques were used to polymerise and study a variety of conducting polymers, focusing on polypyrrole based sensors with differing functionality. The use of different dopant ions was also studied in an effort to optimise the sensitivities of the polymer sensors. The results of electrochemical development and gas testing were used to elucidate the optimal sensor array in relation to the calibration gases used, which was subsequently used in the hybrid device prototype. the conducting polymer sensors did not perform well using solid phase microextraction sampling methods, results using a direct injection method of the headspace showed that the device could discriminate between Pseudomonas aeruginosa and Staphyloccocus aureus.

Bone tissue engineering aims to regenerate the bone structure and therefore recover the functions of bone tissue rather than replacing it alone. Regenerative medicine focuses on using biomaterials as three-dimensional (3D) porous scaffolds, specifically designed to mimic the nature of host tissue and hence to promote cell growth and tissue regeneration. For such purposes, 3D bioactive glass scaffolds are one of the most studied types of scaffolds for bone tissue engineering because of their excellent bioactivity and potential for stimulating osteogenesis and angiogenesis. In the present study stereolithography has been used to fabricate negative moulds for use with the gel casting process to produce porous 3D 70%SiO2-30%CaO2 bioactive glass-ceramic scaffolds with three different pore sizes and identical porosity. A scaffold with 50 vol. % solid loading suspension was successfully manufactured in two different 3D external shapes and three pore sizes. The bioactive glass powder was crystallized at a temperature of 865.5°C. The mechanical behaviour of the scaffolds sintered at 1200⁰C was found to be influenced by pore size despite the similarity in porosity and the scaffold compressive strength decreased, and the failure probability increased, with increasing pore size. This behaviour was found to be consistent with the predictions of Weibull statistics. All three scaffold types exhibited a compressive strength within the strength range of human trabecular bone. The indentation hardness of the scaffold struts was found to be close to that of cortical bone. In vitro investigation of the scaffolds’ bioactivity was achieved through examining changes in the composition of the immersion solution. Biological tests showed that all scaffolds significantly enhanced cell proliferation, deposition of collagen, alkaline phosphatase activity and the expression of osteocalcin with an increasing rate of mineralisation throughout the culture period; this is believed to be due to the action of released ions from the bioactive glass which induces osteoblast cells from their proliferation phase to a mineralisation stage. A 3D sliced scaffold was produced from an assembly of quasi-2D slices to investigate cell behaviour throughout the scaffold. The goal of in vitro studies of the sliced scaffolds with different pore size is to improve the understanding of how scaffold pore size impacts on initial cell attachment, tissue ingrowth and mass transfer through the scaffold. The results confirmed that a scaffold with bigger pore size provided more space for tissue ingrowth and mass transfer throughout the scaffold over long culture periods. The findings suggest that the fabricated 3D 70%SiO2-30%CaO2 bioactive glass-ceramic scaffolds have potential for use in bone tissue engineering applications.

The long term aim of this work is to investigate and develop efficient methods to analyse systems involving biological flows with fluid structure interaction, particularly cerebral aneurysms. Cerebral aneurysms are extremely complex biological structures and this work develops some of the groundwork required to eventually build the capability to efficiently simulate them computationally. This will enable the prediction of a patient specific risk model to aid in surgical decisions on a day to day basis or in large scale studies. Fibrous immersed dynamic structures of cerebral aneurysms seem suited to immersed boundary method simulations. Because of this numerous Immersed Boundary methods are investigated to determine if they present a viable approach, and which of these is best suited. A number of different computational tools have been built and investigated using FFTW, Eigen, OpenMP, and GNU OCTAVE to meet the requirements to allow numerous implicit and explicit approaches to be investigated. These tools are validated and tested for a number of cases. Preliminary work is also presented aimed at generating physically representative numerical models from MRI scans.

Tribocorrosion plays a significant role in the performance and failure of implantable metallic devices. The damage caused by tribocorrosion has been reported previously in several implantable devices such as disk replacements, bone plates, the interface between the fixation cement and the metallic stem in cemented hip arthroplasties and also in the taper-trunnion contact area of hip replacements. The origin of the tribocorrosion processes is produced at the nanoscale as a result of the micromotion between hard single asperities and the metallic components at their contacting interfaces. Hard particles cause deformation and wear of the surfaces and depassivation, opening new metallic areas to corrosive attack. The combination of both processes produces the liberation of metallic ions and metal debris responsible for adverse reactions within the body, causing pain and the need for implant revision. Significant research has been undertaken to understand the wear-corrosion mechanisms at the macro and micro scale. The focus has been on different experimental conditions and the effects of the environment (protein-phosphate contents or pH), the role played by the microstructure of CoCrMo alloys and the contact conditions or the wear mode (sliding, fretting, scratching). However, no work has been undertaken into the interaction of plastic deformation and corrosion mechanisms at the nano-scale. For a better understanding of the combined effects of the corrosion and the deformation processes, a new electrochemistry cell was designed in combination with the nanoindenter systemto simulate a single asperity in contact with a metallic surface. Four CoCrMo alloys with different manufacturing and thermal histories were analysed toobtain a deformation-corrosion map which summarizes their scratch-corrosion performance at the nano-scale. The microstructure, chemical composition and mechanical properties of the Forged, As Cast (AC), As Cast thermal treated (AC-TT) and As Cast with low carbon content (AC-LC) Co based alloys were studied. Grain size, carbide morphology, carbon content and crystallographic phases present were analysed by metallographic preparation, Scanning Electron Microscopy (SEM) observations and Electron Backscattered Diffraction (EBSD) techniques respectively. Hardness (H) and Young’s Modulus (E) were calculated by indentation and nano-indentation techniques. Static corrosion behaviour of the four alloys immersed in 0.9 wt.% NaCl solution was studied using open circuit potential (OCP) and potentiodynamic (PD) polarization experiments to understand the corrosion mechanisms affecting the alloys without the interference of the plastic deformation and to estimate the minimum stabilization time required to reach a steady potential to be used in the nano-scratch corrosion experiments. The deformation processes under fretting, reciprocating sliding and scratch experiments in dry conditions were also characterized by measuring tangential friction forces, coefficient of friction and plastic deformation values. Post experimental surface analysis was performed to analyse the oxide layer formation and the wear scar morphology. The four CoCrMo alloys were tested under several loading conditions using the new electrochemistry cell performing single scratch-corrosion experiments. The results demonstrated that the crystallographic orientation of the grains produced characteristic deformation features. These features were directly linked to therepassivation times and current densities and were governing the deformation corrosion processes at the nano-scale and the liberation of metallic ions. This work establishes a novel experimental technique that gives a better understanding of the deformation-corrosion processes occurring at the nano-scale in CoCrMo alloys. In addition the results obtained will be useful to help interpret the failure mechanisms observed in retrieved implants and improve the design and development of new materials and material pairing selection for future implants. Moreover the technique developed as part of this work can be extended not just to biomedical applications but also to any other applications in the materials science field where passivated metals are used in a corrosive environment and single asperity contacts act to deform metallic surfaces.