Tissue engineering aims to replace diseased or damaged tissue, one approach for which is to implant cells and a scaffold developed in vitro. One of the engineering challenges is the design of the cell-material interface. Cells respond to a wide range of signals from their substrate, which can be used to control cell behaviour and improve the properties of the implant. Of particular interest to regenerative medicine is that substrate properties can influence stem cell self-renewal and differentiation. This thesis aimed to first better understand how substrate properties such as chemical composition and topography affect stem cell behaviour. For this, two substrates were studied: one with micrometre scale topography, and one with varying chemical composition and nanometre scale topography, for their effect on murine embryonic stem cell (ESC) self-renewal and early differentiation. On micrometre scale topography, the first example of central pit formation due to substrate cues was observed, and the substrates mitigated endoderm specification. On nanometre scale topography, ESC early differentiation markers were diminished, and both substrate topography and chemical composition affected cell behaviour. Then attention was shifted to how biological cues from the substrate, such as ligands specific to cell surface receptors, could guide stem cell chondrogenic differentiation. In that study, the changing adhesion requirements of human mesenchymal stem cells were studied in the form of integrin transcript expression and finally, the role of one specific integrin was shown to affect chondrogenic differentiation in vitro. This was the first complete characterisation of integrin expression and the first demonstration of the role of integrin αvβ8 in chondrogenic differentiation. Overall, these results improve the understanding of how stem cells respond to substrate cues, including some of the crucial cues they require in differentiation, and could therefore be used to improve the design of tissue engineering scaffolds.

A critical challenge in designing materials for tissue engineering (TE) is to provide essential cues that can control cellular behaviour and promote tissue regeneration. TE with fibrous scaffolds by using electrospinning is emerging as a major research area in the field of regenerative medicine. This thesis presents the development of novel electrospun fibrous acrylate scaffolds for bone TE. Acrylate fibrous scaffolds were developed by electrospinning photocrosslinkable and low molecular weight acrylate monomers, methyl acrylate (MA) and diethylene-glycol dimethacrylate (DEGMA). Photocrosslinked fibres were successfully produced by electrospinning different MA and DEGMA compositions and post-UV crosslinking. The ability to produce topologically and mechanically diverse fibrous scaffold materials was demonstrated. Varying MA and DEGMA composition affected overall fibre morphology, swelling and mechanics of the fibrous scaffold. An assessment of biological activity of the acrylate fibrous scaffold was performed to evaluate the effect of varying ratios of MA/DEGMA of the fibrous scaffold on the viability of two different cell types, osteosarcoma-derived osteoblastic cells (Saos-2) and mesenchymal stromal cells (hMSCs). The potential of MA/DEGMA fibrous scaffolds to support Saos-2 cell viability and proliferation was demonstrated. However, the considerable increase in apoptosis of hMSCs cultured on both fibrous and flat samples suggested a lower potential of the MA/DEGMA scaffolds to support hMSCs cell attachment and viability. The fibrous scaffolds were immobilized with synthetic peptides utilizing cysteine-functionalized RGD or DGEA peptide sequences in combination with MA/DEGMA monomers and by employing a photoinitiated mixed-mode thiol-acrylate polymerization mechanism. Cysteine-functionalized DGEA and RGD peptides were incorporated efficiently in the synthesized acrylate scaffold. The peptide-conjugated fibrous scaffolds showed increased hMSCs adhesion and viability. Through cell adhesion and soluble peptide competition assays, the bioactivity and specificity of each peptide conjugated to the scaffold was confirmed. Finally, hMSCs cultured on DGEA conjugated scaffolds exhibited the activation of osteogenic differentiation markers, alkaline phosphatase (ALP) and osteocalcin (OCN). The results presented in this thesis strongly suggest the potential of the acrylate fibrous scaffold for bone TE.

Spectral analysis of Doppler signals plays a very important role in non-invasive measurements of blood velocity distributions. Among the various spectral analysis methods available, the fast Fourier transform (FFT) is regarded as a "traditional" spectral analysis tool and is widely used in commercial, clinical, experimental and research equipment. Some drawbacks of this method, however, have imposed limitations on its use in some clinical cases. A numbers of spectral and time-frequency analysis methods have been studied in this dissertation. These include the traditional FFT, the autoregressive (AR) method, the Wigner-Ville distribution (WVD), and the Choi-Williams distribution (CWD). The advantages and disadvantages of each method have been studied and summarised. Efforts have been made to improve the temporal and frequency resolution of the results. New analysis methods such as the WVD and CWD have been interpreted physically, and some of their new properties have been explored. The results have suggested that the heights of the peaks in the AR spectra of narrow-band signals are not necessarily proportional to signal power, and should be used with caution in the context of Doppler signal processing. The results have also shown that it is appropriate to use the WVD or CWD to analyse signals when high temporal resolution is required. In practice, it is easier for the operator to handle the WVD, which usually produces reasonably good results. The above methods have been applied in practice. Considerable software and hardware development has been carried out, and a number of analysers have been implemented for use under different practical conditions. These analysers were also used to compare experimentally the analysis methods mentioned above, and to confirm the results of theoretical analyses. Some of these analysers have found applications in clinical practice.

The ability to reprogram somatic cells toward a stem cell phenotype has created a powerful method of reversing the fate of lineage restricted cells. If the potential of reprogramming is realised we will be able to provide a source of autologous, functional cells circumventing many of the issues associated with donor transplantation and provide a source of stem cells free from the controversy surrounding derivation from embryos. MicroRNA (miRNAs) are an attractive substitute to foreign DNA for induction of reprogramming. This thesis focused on a specific miRNA, miR-145 due to its role as a repressor of pluripotency during embryogenesis. The majority of reprogramming approaches to date have utilised viral vectors to deliver inducing stimuli however a considerable amount of literature strongly suggests that cells reprogrammed using viral vectors are unsuitable from a clinical perspective due to risks associated with insertional mutagenesis and potential immunogenicity. Therefore this investigation sought a non-viral method of delivering inducing stimuli and an electroporation based technique, known as nucleofection was identified. This non-viral delivery method was evaluated in terms of its efficiency and cell viability and was found to be highly effective in its capacity to deliver both plasmid DNA and miRNA inhibitors to human epidermal keratinocytes (hEKs). Nucleofection of hEKs with an antisense oligonucleotide, which effectively inhibits miR-145 was shown to significantly increase the expression of Oct4, Sox2, Klf4 and c-Myc all of which are required for the successful reprogramming of somatic cells. MiR-145 inhibited hEKs were subsequently subjected to modified culture conditions in an attempt to further promote the expression of pluripotency-associated transcription factors. hESC media containing growth factors, known to promote a human embryonic stem cell phenotype, lead to a distinctive change in cell morphology with cells clustering together to form colonies and also further stimulated the expression of pluripotency associated genes. While hEKs did exhibited re-expression of normally silenced pluripotency genes they had not undergone full cellular reprogramming however now presented a more ‘plastic’ phenotype. A direct reprogramming approach was formulated where miR-145 inhibition of hEKs was performed and post-nucleofection cells were incubated using either standard hEK media or growth factor supplemented hESC media for 5 days. Defined osteogenic, chondrogenic or adipogenic differentiation media were then applied to stimulate reprogramming towards ‘tri-lineage’ fates, whilst crucially demonstrating transgermal differentiation of hEKs. A defined neurogenic media was also applied to further validate their differentiation potential and together this provided further evidence for acquisition of a ‘plastic’ phenotype with identification of lineage specific markers following directed differentiation. Taken together this investigation details a novel, miRNA mediated strategy for the direct reprogramming of hEKs in the absence of foreign DNA or viral vectors whilst facilitating re-differentiation into a number of desirable lineages offering significant advances for clinical translation.

For decades, research has been focussed on finding a way to produce artificial blood as a resolution for the insufficient amount of blood components provided by donation and to provide a more transportable alternative with a longer shelf life. Red blood cells (RBCs) are the most common cell type in blood and are responsible for oxygen transport throughout the human body. It is therefore extremely important to find an alternative oxygen carrier whether these are tissue engineered RBCs or a chemically defined oxygen delivery system The study conducted for this thesis was part of a larger project called Redontap, and was aimed to develop a bioreactor for the manufacture of RBCs. During this research to produce RBCs from adult stem cells in vitro, the main goal was to upscale haematopoietic and erythroid cultures. Understanding the biological signals and their temporal magnitude involved in the division, maturation and migration of the CD34+ haematopoietic stem cells (HSCs) and their differentiated progeny would allow for a controlled continuous production of mature blood cells. The differentiation of HSCs into different blood cell types occurs within different bone marrow niches and so mimicry of the erythrocyte niche is likely to result in maximisation of the rate of red blood cell development. Published research provides evidence that peripheral blood mononuclear cells (PBMCs), including CD34- cells, will be advantageous for erythroid maturation. For this thesis, CD34+ cells were expanded within a population of PBMCs on a stromal layer to recreate a niche-like environment. This approach was also utilised with umbilical cord blood isolated MNCs (UBMCs) to compare HSC expansion potential and subsequently efficiency in erythroid maturation was analysed. Whereas the cell output was limited, differentiated cells proved positive for a range of RBC surface markers and haemoglobin content. As part of the aim for upscaling cell culture by translating static cultures to bioreactor processes, bioreactors with volumes varying between 250mL-3L were analysed for cell retention and viability to achieve high cell densities whilst refreshing culture medium, monitoring culture parameters (e.g. pH, dissolved oxygen), and introducing an hypoxia environment for mimicking the in vivo stem cell niche. In general, this research was focussed on improving dynamic culture conditions for generating higher numbers of cultured erythrocytes than so far has been achieved.

Biomechanical motion capture is the process of recording the movements of people or animals. As an analysis tool it offers valuable insight into human motion and is useful to monitor treatment during rehabilitation. The spine in particular receives a significant amount of attention by biomechanical researchers, as spinal health is directly related to the quality-of-life of an individual. Spinal motion capture improves the understanding of the function and vulnerabilities of the spine as a mechanical structure and the analysis of spinal kinematics, in conjunction with spinal loading, offers a method for analysing therapeutic interventions. Numerous motion capture systems and devices are currently available, each with its own strengths and weaknesses. The systems or devices selected by researchers are usually determined by study objectives. For example, a video motion capture system would clearly not be appropriate in a study designed to monitor lower back movements of factory workers. Instead, unobtrusive accelerometry based devices would be more suitable to measure kinematics in a free-living environment. Accelerometry has its drawbacks however. It is limited to only global pitch and roll measurements and requires a subject to make relatively slow movements (i.e. the acceleration component of movement measured by the accelerometer must be significantly smaller than one g). In general, trade-offs exist between accuracy, obtrusiveness, ease-of-use, cost, mobility (degrees-of-freedom) and clinical versus free-living measurements. This thesis proposes an electrogoniometer, which meets many of the above mentioned criteria. The electrogoniometer aims to be accurate yet unobtrusive, low cost (perhaps less than £5 000, compared to the £100 000 price tag of high-end marker-based video motion capture system) and measures high mobility movement (typically the rotation components of a spinal motion segment) and do so within a free-living setting. The electrogoniometer is composed of discrete goniometers, referred to as goniometer-nodes. The goniometer-nodes are chained serially together to construct a multi degree-of-freedom electrogoniometer. The goniometer-nodes consist of mechanical structures embedded with optical sensors, each capable of measuring four degrees-of-freedom (three rotations and one translation). The mechanical structure's articulation is determined by processing the optical sensor data using the principles of triangulation and trilateration. The articulation is measured relative to a local reference frame (i.e. relative to the proximal-end of the node). Since local reference frame measurements are involved, accuracy and precision are important. Poor accuracy and precision will result in measurement errors propagating through the chain. The rotation accuracy is estimated to be better than 2° per axis (which is much less than the typical 5° accuracy of commercial goniometers) and a displacement (translation) accuracy of less than 0.2 mm. Precision is estimated better than 0.5° degrees per axis and 0.1 mm. The device is particularly suitable to measure spinal movement. It is attached to the backof a subject, similar to commercial electrogoniometers. It monitors the spinal kinematics on a continuous basis and transmits the data to a computer via a wireless adaptor. The kinematic data is then available for further analysis. This thesis initially investigates the mechanical and sensor design of the goniometer-nodes. A mechanical composite structure consisting of an universal (two rotations) and cylindrical (one rotation and translation) joint was utilised. Optical emitter-detector pairs were embedded within the structure, and a mathematical model was derived to predict the response of the detectors based upon the kinematic input. A custom instrument was developed to calibrate the nodes. Five nodes were assembled and calibrated, and then chained together to produce the electrogoniometer. The second part of the thesis evaluates the device. Reflective triads were attached to the base of each node within the chain. The device was then manipulated manually and compared against a video motion capture system for accuracy and precision. Analysis of the results showed a local reference frame accuracy and precision of 1.9 ± 1.0° per axis for rotation and 3.5 ± 1.8 mm for translation. The video captured measurements were also compared to the calibration results and proved to be worse than predicted. The cause was traced to the calibration instrument and the measurement method. Although the accuracy and precision specification were not met, it was concluded that the proof-of-concept electrogoniometer demonstrated here has merit as a low-cost motion capture device. The optical measurement method from which the electrogoniometer kinematics are determined, shows promise as a novel kinematic sensing method. It was concluded that with further refinement and improvements of the custom-build calibration instrument, the accuracy and precision requirements can be met. Nonetheless, the concepts and principles have been shown to be valid, and with additional resources, this electrogoniometer can be a viable biomechanical research device.

Conventional bone grafts are fraught with limitations and three dimensional (3D) electrospun fibrous nanocomposites of gelatin and hydroxyapatite (HA) similar to the extracellular matrix (ECM) of bone are viable bone graft substitutes but there is limited research in this area. In this project, fibrous scaffolds of gelatin-HA nanocomposites were fabricated using electrospinning and crosslinked using glutaraldehyde vapour. The microstructural, physical and mechanical properties of the scaffolds were measured and the effects of applied voltage, HA concentration and crosslinking duration on scaffold properties were determined and used to optimise the electrospinning process. Human foetal osteoblast cells were grown on the scaffolds and cell seeding efficiency, cell proliferation, cell viability, alkaline phosphatase activity, collagen matrix synthesis and mineralisation were quantified. Tissue engineered 3D bone grafts were developed by stacking together optimised seeded scaffolds using the three-stack and the four-stack models and also cultured under dynamic conditions in a perfusion bioreactor to improve nutrient and oxygen transport to cells. Mathematical models were developed for nutrient and oxygen transport and cellular response in the scaffolds and layer-by-layer oxygen and cell concentrations were predicted in the 3D bone graft models. Models were developed for cell growth and oxygen consumption rates and their constants were determined and used as input parameters for the mathematical models along with the determined physical and biological scaffold properties in the computer simulations of bone tissue engineering. The scaffolds exhibited a good degree of fibre alignment and both fibre and pore diameters exhibited inverse relationships with applied voltage and HA concentration. The scaffolds possessed reasonable levels of porosity and permeability which were a function of the fibre diameter. Young’s modulus and ultimate tensile strength were functions of fibre diameter, porosity and direction of loading and exhibited proportional relationships with applied voltage, HA concentration and crosslinking duration. Initial cell seeding efficiency was over 90% in all scaffolds with cell proliferation, alkaline phosphatase activity, collagen synthesis and mineralisation all exhibiting inverse relationships with applied voltage and proportional relationships with HA concentration as a result of the concomitant variations in fibre diameter, pore diameter and porosity of the scaffolds. The 25 wt% HA scaffold electrospun at 20 kV had optimum osteogenic, physical and mechanical properties and contained mineralised bone tissue after 18 days of cell culture. Functional 3D bone grafts were obtained with favourable cell proliferation, which improved under dynamic culture, albeit with limited cell migration and a 3D multi-thin-layered stacked bone graft model was proposed based on these findings. Finally, the mathematical models were successfully validated against the experimental cell concentration and migration depth data.

In the work described in this thesis, new techniques to reduce the electron paramagnetic resonance (EPR) irradiation power required in field cycled proton electron double resonance (FC-PEDRI) and to improve the temporal resolution of FC-PEDRI have been investigated. These are the aims of this work. Details of free radicals and basic principles of EPR and EPR imaging are given and the fundamentals of NMR are also described. The principles of NMR imaging (MRI) are summarised and the essential hardware of MRI for the experiments are also described. Details of PEDRI and FC-PEDRI are also described in this thesis. It is widely realised in MRI that a surface coil, or local RF coil, generates very strong magnetic fields in the coil proximity and that the field decreases rapidly with the distance away from the coil. This should be a useful advantage to limit the EPR irradiation within the desired area of the sample. Hence, the use of a surface coil as an EPR irradiation resonator has been investigated. Not only is the EPR irradiation restricted to the area of interest close to the surface coil but a stronger RF field and high SNR are also generated in this region too. The double resonance coil assembly developed here consisted of a split-solenoid coil (for NMR) and a loop-gap resonator surface coil (for EPR). The results confirmed that the enhancement of NMR signal was higher compared to the whole-body birdcage EPR coil used previously and that the enhancement area was restricted within the proximity of the surface coil. Using the technique of rapid imaging in NMR with FC-PEDRI, fast imaging of free radicals has been investigated. A number of fast pulse sequences are presented in this thesis. A snapshot centric-reordered phase-encoding pulse sequence using the technique of population preparation has been studied and employed. With an essential ideal of EPR irradiation applied only once followed by a snapshot NMR imaging pulse sequence, the EPR power is greatly reduced and the temporal resolution is greatly improved, too. However, the experimental results showed that the image quality needed much improvement and this technique is still limited to be used with samples where the longitudinal relaxation time (Ti) is longer than 250 ms. Next, the rapid imaging of free radicals in very short Ti samples equivalent to biological tissue (typically Ti of -150 ms) was investigated. The basis of image artefacts was studied. From the experimental results, it was found that the snapshot NMR pulse sequence must be commenced very soon after the end of field cycling to avoid the signal enhancement decay. To address the problem of image artefacts, the NMR FID data have been analysed. As a result, phase shift correction and amplitude adjustment schemes have been adopted and applied to the FID data. Details of the data correction schemes are given. Experiments employing rapid imaging of free radicals in vivo using FC-PEDRI are presented. The results show that the rapid imaging in vivo using this technique are possible. However, the resulting images are still noisy. More study is required for further refinement of this technique. Experiments using EPR surface coils with rapid FC-PEDRI were also investigated. The preliminary results obtained from biological-equivalent samples show the possibility of the use of this method in in vivo experiments.

Electrical Impedance Tomography (EIT) is a non-invasive imaging method which allows internal electrical impedance of any conductive object to be imaged by means of current injection and surface voltage measurements through an array of externally applied electrodes. The successful generation of the image requires the simulation of the current injection patterns on either an analytical or a numerical model of the domain under examination, known as the forward model, and using the resulting voltage data in the inverse solution from which images of conductivity changes can be constructed. Recent research strongly indicates that geometric and anatomical conformance of the forward model to the subject under investigation significantly affects the quality of the images. This thesis focuses mainly on EIT of brain function and describes a novel approach for the rapid generation of patient or subject specific finite element models for use as the forward model. After introduction of the topic, methods of generating accurate finite element (FE) models using commercially available Computer-Aided Design (CAD) tools are described and show that such methods, though effective and successful, are inappropriate for time critical clinical use. The feasibility of warping or morphing a finite element mesh as a means of reducing the lead time for model generation is then presented and demonstrated. This leads on to the description of methods of acquiring and utilising known system geometry, namely the positions of electrodes and registration landmarks, to construct an accurate surface of the subject, the results of which are successfully validated. The outcome of this procedure is then used to specify boundary conditions to a mesh warping algorithm based on elastic deformation using well-established continuum mechanics procedures. The algorithm is applied to a range of source models to empirically establish optimum values for the parameters defining the problem which can successfully generate meshes of acceptable quality in terms of discretization errors and which more accurately define the geometry of the target subject. Further validation of the algorithm is performed by comparison of boundary voltages and image reconstructions from simulated and laboratory data to demonstrate that benefits in terms of image artefact reduction and localisation of conductivity changes can be gained. The processes described in the thesis are evaluated and discussed and topics of further work and application are described.

Currently, most titanium implant coatings are made using hydroxyapatite and a plasma-spraying technique. There are however limitations associated with the plasma-spraying process including; poor adherence, high porosity and cost. An alternative - the sol-gel technique offers many potential advantages but is currently lacking research data for this application. Hydroxyapatite (HA), fluorhydroxyapatite (FHA) and fluorapatite (FA) have been synthesised by a sol-gel method. Calcium nitrate and triethyl phosphite were used as precursors under an ethanol-water based solution. Different amounts of ammonium fluoride (NH4F) were incorporated for the preparation of the FHA and FA sol-gels. Optimisation and characterisation of the sol-gels was carried out using, X-ray Diffraction (XRD), High Temperature X-Ray Diffraction (HTXRD), Fourier Transform Infrared Analysis (FTIR) and Differential Thermal Analysis (DTA). Rheology and hydrophilicity of the sol-gels showed that increasing fluoride ion substitution caused an increase in viscosity and contact angle. The dissolution (Ca2+ and PO4 3-rates) rates of the fluoride-substituted powders from the sol-gels were considerably lower than that of HA and all rates could be decreased by increasing the sintering temperature. This suggests the possibility of tailoring the solubility of any coatings made from the sol-gels through fluoride ion substitution and increased sintering temperature. A spin coating protocol has been established for coating the sol-gels onto titanium. Increasing the coating speed decreased the porosity and thickness of the coatings. Bond strengths to titanium were investigated. Fluoride substitution and sintering temperature were shown to be important factors. Cellular proliferation studies revealed that increasing the level of fluoride substitution in the apatite structure significantly increased the biocompatibility of the material. The sol-gel technique may be an alternative to plasma spraying for coating titanium implants. Furthermore it may also be suitable for producing HA, FHA and FA as bone grafting materials.