In this dissertation, advanced methods for electroencephalogram (EEG) signal analysis in the space-time-frequency (STF) domain with applications to eye-blink (EB) artifact removal and brain computer interfacing (BCI) are developed. The two methods for EB artifact removal from EEGs are presented which respectively include the estimated spatial signatures of the EB artifacts into the signal extraction and the robust beamforming frameworks. In the developed signal extraction algorithm, the EB artifacts are extracted as uncorrelated signals from EEGs. The algorithm utilizes the spatial signatures of the EB artifacts as priori knowledge in the signal extraction stage. The spatial distributions are identified using the STF model of EEGs. In the robust beamforming approach, first a novel space-time-frequency/time-segment (STF-TS) model for EEGs is introduced. The estimated spatial signatures of the EBs are then taken into account in order to restore the artifact contaminated EEG measurements. Both algorithms are evaluated by using the simulated and real EEGs and shown to produce comparable results to that of conventional approaches. Finally, an effective paradigm for BCI is introduced. In this approach prior physiological knowledge of spectrally band limited steady-state movement related potentials is exploited. The results consolidate the method.

In this Thesis a hierarchical image-based 2D talking head model is presented, together with robust automatic and semi-automatic animation techniques, and a novel perceptual method for evaluating visual-speech based on the McGurk effect. The novelty of the hierarchical facial model stems from the fact that sub-facial areas are modelled individually. To produce a facial animation, animations for a set of chosen facial areas are first produced, either by key-framing sub-facial parameter values, or using a continuous input speech signal, and then combined into a full facial output. Modelling hierarchically has several attractive qualities. It isolates variation in sub-facial regions from the rest of the face, and therefore provides a high degree of control over different facial parts along with meaningful image based animation parameters. The automatic synthesis of animations may be achieved using speech not originally included in the training set. The model is also able to automatically animate pauses, hesitations and non-verbal (or non-speech related) sounds and actions. To automatically produce visual-speech, two novel analysis and synthesis methods are proposed. The first method utilises a Speech-Appearance Model (SAM), and the second uses a Hidden Markov Coarticulation Model (HMCM) - based on a Hidden Markov Model (HMM). To evaluate synthesised animations (irrespective of whether they are rendered semi automatically, or using speech), a new perceptual analysis approach based on the McGurk effect is proposed. This measure provides both an unbiased and quantitative method for evaluating talking head visual speech quality and overall perceptual realism. A combination of this new approach, along with other objective and perceptual evaluation techniques, are employed for a thorough evaluation of hierarchical model animations.

Intravascular ultrasound has, over the past decade, significantly changed the clinical diagnosis and therapeutic strategy of coronary and vascular disease assessment, as it not only allows visualisation of the vessel lumen, but gives a unique view of the pathophysiologic structure of the artery wall. This information is currently unavailable from the universally accepted instrument for artery assessment, angiography, which has on several occasions had its diagnostic accuracy questioned. With intravascular ultrasound, there is the potential to categorise diseased arterial tissue belonging to distinct pathological groups which can ultimately aid in the understanding of individual lesions as well as making a significant contribution to treatment choice and management of cardiac patients. The high resolution image information offered by intravascular ultrasound provides excellent cross-sectional views of coronary artery disease at the level of the disease process itself. This information can be used by the clinician to characterise atherosclerotic plaque composition and vessel wall morphology, both of which are important, in determining the clinical response to the disease condition. However, this visual diagnosis is in general highly subjective due to inter- and intra-observer error. To overcome the short comings inherent in the visual assessment of intravascular ultrasound images, texture analysis was used to assess plaque in regions of interest identified by a clinician. In the two dimensional images produced by intravascular ultrasound, texture is perceived as homogeneous visual patterns representing the surface composition being imaged. Since every tissue sub-group has its own texture, verified from histological analysis, it can be used as a means of characterising it. In this thesis, the findings of applying texture analysis techniques to 30 MHz intravascular ultrasound data, gathered in vitro, to assess its potential in quantitative coronary plaque characterisation are presented. Histo-pathological analysis was used to form a gold standard based upon clot composition, from which the results were verified. The ultimate aim of the work was to determine a reliable protocol based upon textural analysis for assessing plaque composition in vivo. Textural properties, in the form of features, were calculated for regions of interest using first-, second- and higher-order statistics. These were found to be computationally expensive and in certain instances produced duplicate, and hence redundant, information. Feature selection was used to increase the computational efficiency of the algorithm by optimising the feature set. In a further attempt to overcome the weaknesses of the aforementioned techniques, fractal texture analysis was used to obtain textural information on regions of interest. Fractals proved useful in describing the texture of these areas by a single measure. This measure, the fractal dimension, described the degree of irregularity in the surface texture. A new method is proposed for classifying arterial plaque which relies on a combination of the two powerful techniques previously mentioned, statistical and fractal texture analysis. The results presented show the ability of the texture analysis techniques used to discriminate certain tissue sub-groups. Limited success was achieved for the analysis on the atherosclerotic plaque groups studied, however, the approach adopted significantly discriminated the three types of clot composition studied: plasma; white thrombus; and red thrombus.

The electrophysiological observation of neurological cells has allowed much knowledge to be gathered regarding how living organisms are believed to acquire and process sensation. Although much has been learned about neurons in isolation, there is much more to be discovered in how these neurons communicate within large networks. The challenges of measuring neurological networks at the scale, density and chronic level of non invasiveness required to observe neurological processing and decision making are manifold, however methods have been suggested that have allowed small scale networks to be observed using arrays of micro-fabricated electrodes. These arrays transduce ionic perturbations local to the cell membrane in the extracellular fluid into small electrical signals within the metal that may be measured. A device was designed for optimal electrical matching to the electrode interface and maximal signal preservation of the received extracellular neural signals. Design parameters were developed from electrophysiological computer simulations and experimentally obtained empirical models of the electrode-electrolyte interface. From this information, a novel interface based signal filtering method was developed that enabled high density amplifier interface circuitry to be realised. A novel prototype monolithic active electrode was developed using CMOS microfabrication technology. The device uses the top metallization of a selected process to form the electrode substrate and compact amplification circuitry fabricated directly beneath the electrode to amplify and separate the neural signal from the baseline offsets and noise of the electrode interface. The signal is then buffered for high speed sampling and switched signal routing. Prototype 16 and 256 active electrode array with custom support circuitry is presented at the layout stage for a 20 μm diameter 100 μm pitch electrode array. Each device consumes 26.4 μW of power and contributes 4.509 μV (rms) of noise to the received signal over a controlled bandwidth of 10 Hz - 5 kHz. The research has provided a fundamental insight into the challenges of high density neural network observation, both in the passive and the active manner. The thesis concludes that power consumption is the fundamental limiting factor of high density integrated MEA circuitry; low power dissipation being crucial for the existence of the surface adhered cells under measurement. With transistor sizing, noise and signal slewing each being inversely proportional to the dc supply current and the large power requirements of desirable ancillary circuitry such as analogue-to-digital converters, a situation of compromise is approached that must be carefully considered for specific application design.

A collaboration between Bracco Imaging S.p.A. and Durham University allowed the work described in this thesis on the design and synthesis of new contrast agents for MRI. Significant enhancements in the relaxivity of contrast agents for MRI can be gained by increasing the complex rotational correlation time ((_TR)). Incorporating a Gd(^111) ion within a ligand structure possessing a suitably large dendritic framework, inspired the first part of this project. Thus, the periphery of a Gd-DOTA derivative was adorned with carbohydrate containing wedges. The symmetry of the mono- aqua tetra-substituted structure places the gadolinium-water vector at the centre of any tumbling motion, allowing a coherent tumbling of the macromolecule and an optimization of its rotational correlation time. The carbohydrates ensured high water solubility and favoured a large second sphere hydration contribution to the relaxivity. An increase in the hydration around the metal centre and a rapid exchange of the water molecules with the bulk solvent can also significantly increase the contrast agent efficacy, by efficiently transmitting the paramagnetic effect from the Gd(^111) to the solvent. In a second part of the work, the development was undertaken of diaqua systems based on the seven-membered heterocycle 6-amino-6-methyl- perhydro-1,4-diazepin (AMPED). The three N-positions were substituted with different phosphinate and carboxylate groups and lanthanide complexes (Eu(^111), Gd(^111), Yb(^111)) prepared and studied by multinuclear NMR methods The alkylation of the amino groups with chiral 1,5-dicarboxylate pendant arms led to complex diastereisomers, possessing different water exchange rates. The individual water exchange rates of each isomer were determined, and differed by a factor of six. Furthermore, the periphery of the isomer possessing a faster water exchange rate was adorned with carbohydrate containing wedges, and the relaxation properties studied.

Introduction The majority of primary and secondary liver tumours are inoperable. „In situ‟ thermal destruction techniques such as radiofrequency, microwave ablation and cryotherapy have been employed to treat these inoperable tumours. Despite recent advances in these technologies, large and peri-vascular tumours still suffer from a relatively high recurrence rate post ablation. This is thought to be due to the loss of thermal energy to surrounding vasculature, known as the heat sink effect. The aim of this project was to investigate the effect of surrounding vasculature on ablation morphology and success and compare the three most popular ablation modalities. Methods Standard sized ablations were created in rat liver at various distances from the hilum with all three methods. At various time points, tissue samples were retrieved and underwent histological (H&E) and immuno-cytochemical (Hsp70 and Caspase 3) staining in order to assess lesion evolution and the effects of surrounding vasculature on ablation completeness. Results All rats survived. The greatest amount of activity was seen in the transition zone. H&E and immuno-cytochemical analyses of lesion evolution discovered previously unreported cellular changes, particularly in the transition zone. Cryotherapy ablation seemed to be the most irregular and unpredictable of the three. Radiofrequency ablation was uniform but showed evidence of extra-lesional apoptosis and perivascular cell survival in addition to Hsp 70 activity in the transition zone that was affected by surrounding vasculature. Microwave ablation seemed to be influenced least by surrounding vasculature and formed the most uniform lesion with very little extra-lesional collateral damage. Conclusion The success of ablation is dependent upon the adjacent blood vessels and microwave ablation seemed to form the most predictable burn and be least affected by surrounding vasculature compared to radiofrequency. Cryotherapy should not be used as first line treatment to treat unresectable liver tumours. In addition the exact role of HSP 70 on the fate of cells in the transition zone, and the subsequent final ablation size and morphology is yet to be determined. Larger ablations in larger animal models may help answer some of these questions.

The need for continuous noninvasive arterial blood pressure (ABP) monitoring from an artery closer to the heart (i.e. the ascending aorta) has led to the research and development work presented in this thesis. Clinical applications of continuous ABP waveform include assessments of cardiac function, cerebral autoregulation, autonomic function, arterial elasticity, physiological measurements in aerospace research, and also monitoring in anaesthesia and critical care. The superficial temporal artery (STA) was chosen as the measurement site and the measurement technique was the arterial volume clamping with photoplethysmography (PPG). The optoelectronic circuitry to measure PPG is contained in a specially designed probe placed over the STA and kept in place with a lightweight aluminium head frame. The complete prototype device (STAbp) also includes original designs for the pneumatic, electronic, signal processing, control and display sub-systems. A self-calibration feature that regularly updates the PPG reference level (Setpt) was also included to ensure accurate continuous ABP recording. The performance of the STAbp was compared against the Finapres®. Five parameters were evaluated: resting ABP (agreement, signal bandwidth, frequency response and magnitude squared coherence, and assessment of drift) and ABP dynamic change during isometric handgrip exercise. The agreement of resting ABP gave bias (SD) of -23.1 (15.05), -10.8 (13.83) and -12.4 (12.93) mmHg for systolic, mean (MAP) and diastolic pressures respectively. Further investigations were carried out to understand factors that can affect the accuracy of ABP measurements, notably the sensitivity of ABP to perturbation of the Setpt. Also, differences between the external compressing pressure at the PPG peak pulsation amplitude and the MAP were found to be normally distributed with mean (SD) of -4.7 (5.63) mmHg. In conclusion, it is demonstrated that the new STAbp device has great potential as a new tool for a wide range of clinical and research applications which require continuous ABP waveforms.

No description available.

It is believed that irradiation interacts with biological tissues to break or modify the DNA, which is the molecule contained in the nuclei of cells that carries all the relevant information for the organism. As such, radiation is dangerous for individuals; however, its properties can also be used in medicine, e.g.. in cancer treatments. Nevertheless, the exact mechanisms of cellular response to radiation are not fully understood yet, especially for low doses (below 50 cGy), where non-targeted effects, i.e. that do not involve only the interactions radiation-DNA, are taking place. In order to deepen the knowledge of those non-targeted effects,a.computer model of a population of cells irradiated in vitro was written, taking into account the phenomena in the low dose domain. As a start, two non-targeted effects were studied, the bystander effect and the low dose hyperradiosensitivity. The program was written in C++ and the technique of the cellular automaton was used. The clonogenic assay was reproduced; cells were seeded in a dish and if the colony they formed after a given period of time was bigger than 50 cells, the seeded cells were assumed to have survived. The direct effect of radiation was calculated by the traditional linear quadratic model and in addition cells were subjected to the bystander effect. Some simulations were run in the case of two cell lines, the hamster cell line V79 and the glioma cell line T98G. The results show that the bystander effect is unlikely to be limited to one period of the cell cycle, but that the low dose hyper-radiosensitivity and the bystander effect could be the same phenomenon. This work also suggests that the bystander effect may be significant after low doses of conventional radiotherapy. Such a model represents a very useful tool for solving problems that at the moment cannot be investigated experimentally.

No description available.