Unicompartmental knee arthroplasty (UKA) has been gaining popularity in recent years due to its perceived benefits over total knee arthroplasty (TKA), such as greater bone preservation, reduced operating-room time, better post-operative range of motion and improved gait. However there have been failures associated with UKA caused by misalignment of the implants. To improve the implant alignment a robotic guidance system called the RIO Robotic Arm has been developed by MAKO Surgical Corp (Ft. Lauderdale, FL). This robotic system provides real-time tactile feedback to the surgeon during bone cutting, designed to give improved accuracy compared to traditional UKA using cutting jigs and other manual instrumentation. The University of Strathclyde in association with Glasgow Royal Infirmary has undertaken the first independent randomised controlled trial of the MAKO system against the Oxford unicompartmental knee arthroplasty - the most common manual UKA used in the UK. This thesis investigates the results from a total of 51 patients (23 Mako, 28 Oxford) that underwent a one year post-operative biomechanical assessment. The assessment analysed the biomechanics of these patients performing walking tasks, stair navigation, sit to stand and deep knee lunges using a 3-dimensional, 12 camera motion analysis system (Vicon Motion Systems, Oxford, UK). 3 month post-operative X-rays confirmed that the implant alignment in the Mako group were significantly more accurate than the implants in the Oxford group. Motion analysis showed that during level walking the Mako group achieved a higher knee excursion during the highest flexion portion of the weight bearing stage of the gait cycle (18.6°) compared to the Oxford group (15.8°). This difference was statistically significant (p-value = 0.03). When compared to normal patients the Mako group's knee excursion values were comparable with normal healthy knees, however the Oxford group had significantly lower knee excursion angles at this point. Even though there were some differences seen in the two groups with motion analysis, these factors did not necessarily correlate with better perceived patient function when the knee function scores were compared against the knee excursions. Therefore it is still unclear if improved implant alignment and better knee motion directly correlate with improved function.

Image registration is an important aspect in all computer assisted surgeries including Neurosurgery, Cranio-maxillofacial surgery and Orthopaedics. It is a process of developing a spatial relationship between pre-operative data, such as Computed Tomography (CT) scans or Magnetic Resonance Imaging (MRI) scans and the physical patient in the operating theatre. Current image registration techniques for Computer Assisted Orthopaedic Surgery (CAOS) in minimally invasive Unicompartmental Knee Arthroplasty (UKA) surgery are invasive, time consuming and often take 14-20 minutes and are therefore costly. The rationale for this study was to develop a new operating theatre compliant, quick, cost effective, contactless, automated technique for image registration during CAOS based on an accurate rigid body model of the ends of the exposed knee joint, produced using 3D laser scans taken intra-operatively by a Laser Displacement Sensor. Bespoke automated 3D laser scanning techniques based on the DAVID Laserscanner method were developed and were used to scan surface geometry of the knee joints in cadaveric legs. The laser scanned knee joint models were registered with the pre-operative (MRI/CT) models and the deviations were evaluated. Furthermore, trends in the deviations were studied along with a supportive validity study. Results indicated that the laser scanner can repeatedly produce accurate 3D models of the human tibio-femoral joint in the operating theatre. This study has provided a proof of concept for a new in situ automated shape acquisition and registration technique for CAOS with the potential for providing a quantitative assessment of the articular cartilage integrity during lower limb arthroplasty.

Spinal cord injury results in severe physical disability and a wide range of progressive medical complications. The main challenge for clinicians and neuroscientists is to develop methods for enhancing recovery after spinal cord injury. New researches have demonstrated that robotic or manual assisted treadmill training can have long lasting positive effects on the recovery of locomotion in incomplete SCI human patients. By moving limbs and progressively modifying body weight support, the patterned sensory information arising from the robotic or manual guidance of movement is considered to increase the potential for the gait recovery. Commonly, deficits in walking in incomplete spinal cord injured patients are often revealed as deficits in ankle control. Accordingly, it is believed that successful recovery of stepping requires a degree of sparing in sensory and motor pathways that subserve ankle control. We therefore have begun experiments that examine ways to facilitate activation in pathways that influence the ankle joint control and that can be used within the context of body weight support rehabilitation programs. The work focused on developing a system for vibratory stimulation of the foot sole that can act as a surrogate stimulus for ground contact and also we studied the physiological effects of vibration in spinal and supraspinal levels. Findings in this thesis demonstrated that short periods of foot vibrotactile stimulation can produce measureable effect at cortical and spinal level in normal subjects. The findings suggest that activation of foot mechanoreceptors using localized vibrotactile stimulation interact with spinal inhibitory control mechanism contributing to the control of locomotion in human. This type of stimulation will most likely can have practical benefits for normalising gait and restoring reflex modulation during gait training. Finding in this study showed that an insole device can make this happen and can be used in gait training of SCI subjects.

Until now, natural thumb control mechanisms are lacking in the upper limb prosthetic development. This lag is due to the complex anatomy of the musculature of the hand making the upper limb prosthetic research a very complicated area. Several studies have attempted to better our understanding of the neural control of the hand. With applications including clinical rehabilitation, surface-Electromyography (sEMG) has been steadily improving the knowledge in this area, albeit still very limited. The development of high-density sEMG (HD-sEMG) has drastically increased the sensitivity of EMG techniques. Despite this research effort, there are significant gaps in the field. Furthermore, current data analysis is almost exclusively performed off-line and so, neurally controlled prostheses are limited to research labs and are not a clinically viable technology. Therefore, it is evident that new technologies are required to understand the dexterity of the human hand for prosthetic control. A common theme across the different hand-prosthetic developers is not to have mechanisms to drive the thumb based on muscular contractions. Due to the lack of intuitiveness for an amputee to operate the prosthetic device, it requires several highly demanding training sessions between the patient and the prosthetist. These sessions are oriented for the amputee to be able to control in duration and magnitude, the contractions of the chosen muscles to drive the motors of the prosthesis. In this research, the muscle activity from the forearm is identified and correlated with specific hand movements leading to improve the commercially available myoelectric transradial prosthetics. This was achieved through the understanding of sEMG patterns related to differentiation of thumb op position to different fingers. The acquired signals were investigated based on time-domain analyses (i.e. amplitude signal analysis, root-mean square values, statistical analyses), followed by a joint time and frequency-domain analyses (i.e. coherence estimate and cumulant analysis). Finally, unsupervised machine learning techniques were applied aiming to differentiate the different sEMG patterns during the different thumb opposition. This differentiation leads to a better understanding with regards to prospective controller mechanisms aiming to develop new prosthetic devices enhancing the experience of transradial amputees with the use of their prosthetic.

The brain is a fascinating machine that is fundamental to our existence as conscious individuals. This is highlighted during neurological disorders that can have a devastating impact on the sufferer's aptitude and quality of life. There is much which is not yet understood about what bhappens during neurological disorders including the changes which occur at a cellular level that affect synaptic communication between neurons. One method of studying these synaptic connections and how they change during disorders is through in vitro neuronal cell cultures which are a valuable tool for investigating cellular mechanisms. Recently, microfluidic techniques have enabled new methods of patterning cells in vitro and can provide precise control of the extracellular environment. Compartmentalised devices have been created that allow for certain characteristics of neurological disorders to be modelled in vitro. However, current methods of applying drugs to neuronal network in such devices are often performed manually which can limit their value as it is impractical to switch between multiple solutions. In this thesis, a method is initially developed for quantifying the synaptic comunication that occurs between functionally connected neural networks that are held in isolated environments. This was investigated using primary hippocampal neurons grown in a compartmentalised device. One sub-network of neurons was chemically stimulated while both presynaptic and postsynaptic responses were observedsimultaneously using Ca²⁺ imaging. Additionally, to address the currently limited methods of altering the extracellular environemtns in neuronal microfluidic devices, a microfluidic perfusion system was developed that can switch between multiple solutions. This was applied to compartmentalised neural networks while their cellular activity was monitored using Ca²⁺ imaging. Overall, the methods developed here can be used to study neurological mechanisms in a controlled manner and have the potential to be used in the screening of novel drugs and therapeutics targeted at neurodegenerative disorders.

Hypoplastic Left Heart Syndrome (HLHS) is a rare congenital heart disease characterised by the underdevelopment of the left sided structures of the heart, compromising systemic blood supply. The Hybrid Procedure is a palliative repair that delays cardiopulmonary bypass surgery and allows the opportunity for left ventricular growth and biventricular repair. The ductus arteriosus is stented open via catheter, which allows the right ventricle to supply the systemic circulation. In order to balance the pulmonary-systemic flow ratio, branch pulmonary arterial bands are surgically placed. Currently, banding (and stent) dimensions are based on surgical experience, intuition and limited Doppler measurements. In mathematically modelling the Hybrid Procedure, it is possible to optimise the dimensions based on haemodynamic and ventricular data. These simulated results are often difficult and invasive to measure clinically. Due to the broad spectrum of abnormalities observed in HLHS, creating patient-specific models is an area for development. Therefore a thorough investigation of routinely collected clinical data was undertaken, assessing the potential collaboration between biomedical engineering and clinical protocols. A lumped circulation model of the post-Hybrid circulation was produced and clinically validated following novel investigation. An external band diameter of 3 mm was optimal, with 3.5 mm appropriate for larger patients. A patient-specific three-dimensional geometry was constructed and virtual surgery performed for a range of band diameters for steady state analysis. Boundary conditions were determined using matching patient-specific and literature data. This model was coupled to the lumped circulation model in a multiscale model. This highlighted the conflict of definition between internal and external diameter band dimensions. It was shown that the 2 mm internal band diameter was optimal. Regarding patient-specificity, it was demonstrated that current clinical practices are not conducive to mathematical modelling with many steps required in the processing of data. The quality of the data is suboptimal and will require multidisciplinary cooperation for future improvement. Due to the incompleteness of the data sets and the inconsistent data collection, full patient-specificity and predictive modelling was not achieved.

Deep-brain stimulation (DBS) is increasingly being pursued as a treatment option for a range of neuropsychiatric disorders. When assessing its potential for the treatment of schizophrenia, related animal studies are often limited by the range of measurements that the equipment is capable of. In animals, the combination of DBS with various recording modalities such as local-field potential (LFP) recording has traditionally required complex and time-consuming laboratory setups. Furthermore, the elucidation of neural activity underpinning rodent behaviour has traditionally been hampered by the use of tethered systems and human involvement. Taken together, novel tools and techniques are required to drive forward DBS research in this area. In this study, two miniature wireless devices were developed for electrophysiological recording and stimulation in freely-moving rodents. The performance of one of these devices was verified in an open-field chamber, in which high-frequency (100Hz) st imulation was delivered bilaterally into the anterior thalamic nucleus at a range of current intensities (20(So(BA, 100(So(BA) and pulse-widths (25(So(Bs, 100(So(Bs, 200(So(Bs). LFP recordings were made bilaterally in the fronto-hippocampal brain regions. Not only was the recording/stimulation device able to successfully correlate electrophysiological recording and stimulation with animal behaviour (via video tracking), but a transient velocity increase of the animals was observed following stimulation at the higher current setting (100(So(BA). The effect of fimbria-fornix (FF) DBS (at 130Hz, 30(So(BA, 90(So(Bs) was then studied in a rodent disease model relevant to schizophrenia, using a spatial working memory paradigm inside a T-maze. Fronto-hippocampal LFP was recorded bilaterally, and was subsequently correlated to the rat's position using synchronised video-tracking. Notably, rat gamma-frequency LFP was found to be increased in all brain regions following an acute administration of the NMDA receptor antagonist phencyclidine (PCP; 3 mk.kg-p1s i.p.), which had persisted throughout the duration of the recording session. Furthermore, rat hippocampal theta-frequency activity was transiently elevated following a 30-second period of FF-DBS, which was carried out during the intra-trial delay period of the task. Finally, the use of FF-DBS throughout the task training sessions highlighted a (non-significant) tendency for rats to reach criterion performance faster than their sham-stimulated counterparts, highlighting the FF as a potential DBS target to consider with regards to disorders that affect learning and memory. The data presented in this study highlights a) the successful design and application of novel device technologies for enhancing the range of measurements in animal-related DBS studies, and b) the effects of FF-DBS in a rodent model relevant to schizophrenia, and its implications in the treatment of this disorder.

Transducer arrays operating above 15 MHz enable real time high resolution imaging of tissue, capable of resolving features below 200μm. Clinical applications such as oncology and gastroenterology could significantly benefit from the improved resolution for high frequency ultrasound (HFUS) characterization of tissues. However, this is presently challenging due to the limited penetration depth of HFUS and limited access. Since the device dimensions scale with imaging wavelength, it becomes feasible to integrate HFUS arrays with interventional tools such as biopsy needles. Although there are many design and fabrication challenges associated with incorporating transducers with interventional tools such as biopsy needles, it creates opportunities for timely and accurate characterisation of tissue, leading to in-vivo pathology. This study reports progress in the development of fabrication processes for miniature linear arrays suitable for integration with biopsy needles. While patterning high frequency transducer arrays based on piezocomposites has been shown to be feasible, there remain many challenges to miniaturize the interconnect and cabling of an ultrasound probe suitable for in vivo pathology. Novel packaging techniques for integrating an ultrasound array into a needle were developed. Wafer scale fabrication was adopted to reduce the overall cost of fabrication. Microfabrication and precision micromachining processes were developed to overcome the technical challenges in fabricating miniature arrays operating up to 25 MHz. Array elements are defined by precision dicing and the necessary external flex circuit cabling was fed through the needle. A flexible printed circuit is connected to back surface electrodes using low-temperature bonding methods. A flex circuit connected to the 1-3 piezocomposite was patterned with 60 μm pitch to define array elements suitable for a 25 MHz linear array. The polyimide flexible printed circuit, with fine pitch traces, was twisted into a helical structure so that it can fit within the core of the biopsy needle and permit large numbers of elements and electrode traces. The spiral-helical flexi-circuit design was developed as a way to fit multiple conductive tracks into a needle. The definition of fine-pitch conductive tracks on polyimide polymer was achieved using dry-film photoresist and the application of a megasonic transducer to provide agitation and small bubbles for copper etching. Investigation and evaluation of low temperature bonding methods was undertaken. This overcomes the problem of using high temperature methods on the temperature sensitive single crystal materials. Bonding techniques such as ultrasonic bonding and magnetically aligned anisotropic UV curable epoxy were investigated. A Resolution integral was applied to simulated beam plots as a way of evaluating transducers at a design stage. This considers the ultrasound beams and a measure of the beam at -6 dB is taken as the lateral resolution. This is measured over the depth of field. A transducer with a higher resolution integral would have a narrow beam over a long distance The process was validated with a single element transducers made from fine-scale single crystal composites involving PMN-PT and Manganese doped Lead Indium Niobate-Lead Magnesium NiobateLead Titanate (Mn-PIN-PMN-PT). These were fabricated using the conventional dice and fill method, and incorporated into needles and tested. These composites had pitches as small as 50 μm with kerf of 18 μm. Images were generated using these transducers. Arrays operating at 5 MHz and 15 MHz were fabricated. The fabrication process development and testing demonstrated the feasibility of a linear array integrated into a biopsy needle. The extension of the fabrication processes to higher frequency arrays.

With mortality rates of both men and women increasing considerably over the last decade from liver disease, along with the number of patients on the active liver transplant list in the UK on the increase, new diagnostic tools would be welcomed by hepatologists. This trend is possibly mirroring the increase in patient numbers with alcoholic liver disease (ALD), and non-alcoholic fatty liver disease (NALFD). The aim of this project was to find a suitable biomarker for the early detection of liver disease. One such biomarker that was proposed was insulin-like growth factor-binding protein-5 (IGFBP-5). IGFBP-5 is known to be involved in the wound healing response in epithelial tissues, and its expression is switched on in fibrosis of the lung. IGFBP-5 induces epithelial cell senescence, an important component in the failure to resolve healing which results in a compensatory fibrotic response. IGFBP-5 has also been identified as a tumour marker gene of interhepatic cholangiocarcinoma, a type of cancer of the bile ducts in the liver. This project aims to determine whether IGFBP-5 could be used as a biomarker for detecting liver damage and fibrosis. Several models of liver disease were developed to determine if IGFBP-5 could potentially be a suitable biomarker using monolayers of primary rat hepatocytes cultured on collagen I coated tissue culture plastic dishes. Firstly, IGFBP-5 release into the culture medium from primary rat hepatocytes cultured over time was investigated along with models of oxidative stress, alcoholic liver disease (ALD), and non-alcoholic liver disease (NALFD). It was determined that the expression of IGFBP-5 over time in culture was increased, suggesting that it may have potential as a biomarker of dedifferentiation. Treatments with menadione, hydrogen peroxide, or ethanol and its primary metabolite acetaldehyde were used to unravel the story on oxidative stress. However, IGFBP-5 was undetectable in the culture medium after chronic treatment with each of the compounds listed. An in vitro model of NALFD was developed at the University of Edinburgh. The model was developed with the ability to induce either enhanced or minimal ROS formation. The in vitro study showed promising results, demonstrating that, with time in culture, in both models, an increase in IGFBP-5 expression was detected. Following on from the in vitro study, a patient study was undertaken to determine if patients with various types of liver disease had any changes in their circulating IGFBP-5 levels. After extensive work, IGFBP-5 could not be detected in human serum using a commercially available kit. This was thought to be due to the human serum interfering with the assay kits ability to detect IGFBP-5. IGFBP-5 does have potential as a biomarker of early liver disease, showing promising results from two models; de-differentiation and NAFLD. Further work into its involvement in liver injury or recovery mechanisms could determine it as useful biomaker in the hep atology clinic.

Micromanipulation tools are a valuable component of biomedical research. A fundamental example is in the study of the complex pathways involved in disease progression in which controlling immune cell interactions at a single-cell level is crucial to the discovery of new treatments. Current cell manipulation tools employ a wide range of mechanisms however, future systems must be geared towards miniaturisation to provide portable and convenient systems which are simple and cost effective for use in research. In this thesis, the use of micro-LEDs as a compact illumination source in optoelectronic tweezers systems is explored. The emerging technique of optoelectronic tweezers uses light patterns to generate electric field gradients to trap and manipulate single cells. Micro-LEDs offer an advantageous alternative to current illumination sources used in this technique, and indeed other light-based micro-systems, in terms of a compact design and control system, low cost and the potential for integration with other micro-systems. Initially, single-cell trapping and fluorescence imaging of immune cells is demonstrated using a micro-LED projection system in which the size of the imaged pixel array was reduced to better match the scale of cells. Using this system, individual cells were trapped and the velocity profile at varying applied voltages and the trap profile were measured. Fluorescently labelled cells were identified in a mixed population through micro-LED excitation and a common indicator of cell activation (calcium fluxing) was also monitored over time showing the combined capabilities of this system. The creation of a novel, integrated micro-LED/OET device for the manipulation of live cells in a compact format is then reported. In this system, the direct integration of a micro-LED array with an optoelectronic tweezers chamber was achieved where cells were successfully manipulated. In addition, interesting combined field effects were observed and potential future developmental prospects were identified.