This project is conducted in collaboration with Cardiff and Vale University Health Board’s (CVUHB) Rehabilitation Engineering Unit (REU).
In our research we have investigated new techniques to improve methods for postural assessment of patients with severe neuro-musculoskeletal conditions. During routine postural assessment, healthcare professionals use anthropometric measurements to infer internal musculoskeletal configuration and inform the prescription of custom contoured seating tailored to individual needs.
However, current assessment process is time consuming and does not readily facilitate communication of musculoskeletal configuration between healthcare professionals nor the accurate recording and comparison of changes over time (that, in turn, would indicate the success or otherwise of interventions thus providing a measure of clinical outcome).
In order to address this unmet need we developed a 3D interactive Digital Human Model (DHM). As an input for the DHM the location of the anatomical landmarks is obtained using an innovative approach of ultrasonic localisation with ultrasound sources as transmitters and miniature microphones as receivers embedded in the glove worn by a clinician.
Our research has produced new insights into the way patients with musculoskeletal deformities are diagnosed, which resulted in several changes proposed towards a new clinical protocol for postural assessment.
These changes include improvement in:
The preliminary results from the clinical trial has demonstrated that the new technology can drive improvements in health and well-being of patients by enabling more accurate postural assessment, better informed clinical decisions and reducing the time patients spend in a clinic.
At least 1000 patients in Wales (20,000 across the UK) should benefit from this technology, representing an annual saving in staff time of approximately £600,000 in Wales (£12m across the UK).
This work is supported by the Welsh Government’s ESF-Funded Knowledge Economy Skill Scholarship with the Cardiff and Vale University Health Board’s Rehabilitation Engineering under Grant MAXI 20422.
Cervical cancer is preventable if effective screening measures are in place. Pap-smear is the commonest technique used for early screening and diagnosis of cervical cancer. However, the manual analysis of the pap-smears is error prone due to human mistake, moreover, the process is tedious and time-consuming. Hence, it is beneficial to develop a computer-assisted diagnosis tool to make the pap-smear test more accurate and reliable. This research describes the development of a tool for automated diagnosis and classification of cervical cancer from pap-smear images.
The interest and use of virtual reality is growing every day as more fields and industries are adopting the technology. A first of its kind survey in the UK, that surveyed more than 100 lecturers, researchers and learning technologists at universities and colleges, has shown that 82% of all respondents said they were very interested in their institution making more use of AR and VR technologies in future (rated 4 or 5 on a scale of 1 to 5).
Our research is focusing on the novel approach for delivering lectures through the use of VR technology. Our main interest is looking into how teaching content can be adapted for a VR delivery and what the changes that need to be made to the said content.
This research has potential for improving overall student engagement in lectures and allow for better understanding of the taught subject. The finding will be used to inform the future VR content development and delivery.
This aim of the project, conducted in collaboration with the Clinical Bioinformatics team, Cardiff and Vale University Health Board (CVUHB) and the Institute of Dermatology, University Hospital of Wales, Cardiff, is to develop a machine learning and advanced image processing algorithm which would lead to new approaches in skin monitoring and early melanoma detection.
AI systems have been highly successful in many applications including face recognition, autonomous driving, image classification, or medical diagnosis, particularly when problems can be expressed as data classification or pattern recognition tasks.
However, AI systems particularly deep learning methods often turn out to be “black boxes,” which create significant challenges in terms of interpreting a predictive result or verifying the accuracy of diagnosis.
In medical diagnosis, for example, one typically seeks not just an answer or an output but also an explanation or structuring of evidence used to support such a prediction. It has been reported that at least 20% of NHS funding is inappropriately spent because too many patients are referred to hospital for treatment which they could have received in a community setting.
This project can play a role in bridging the gap between community-based, primary care and hospital-based secondary care. By reducing the number of referrals and increasing the number of patients being managed in the GP setting this technology can play a significant role in improving both patient care and cost effectiveness.
This work is supported by the Welsh Government’s ESF-Funded Knowledge Economy Skill Scholarship with the Cardiff and Vale University Health Board’s Rehabilitation Engineering under Grant MAXI 21430.
Our researchers collaborated with scientists and clinicians from around the UK to develop a non-invasive infrared, Diabetic Foot Ulcer Prevention System (DFUPS).
The consortium was awarded a grant from National Institute for Health Research (NIHR) Invention for Innovation programme, to develop and validate the device.
In total, in England and Wales, an estimated £14 billion pounds or 10% of the NHS budget is spent a year on treating diabetes and its complications. With the cost of treating complications representing the much higher cost (Diabetes.co.uk 2019).
An increase in skin foot temperature is predictive of neuropathic ulceration in people with diabetes. The rise in local skin temperature can be measured by infrared thermometry. However, this technique is not widely available and at present, mainly single spot hand-held thermometers are used in clinical practice.
These thermometers measure temperature only above the scanned area and do not provide information about the surrounding areas. Therefore, a thermal image of the feet will allow full examination and early detection of areas at risk of foot ulceration.The Diabetic Foot Ulcer Prevention System (DFUPS) was evaluated in a study that observed the temperature distribution in the feet of healthy volunteers assessed with thermal imaging (DFUPS) and spot thermometry as part of a multicentre clinical trial (N. L. Petrov et al 2018).
This study confirmed that the DFUPS device can be used to define the thermal pattern of healthy feet. Both instruments showed that the temperature distribution is similar between the right foot and left foot in the feet of healthy volunteers and these data can be used as normative values when planning studies in people with diabetes.
The success of the study led Dr Peter Plassmann to establish Thermetrix, a medical device company based in South Wales. The company produce ‘Podium’, a foot ulcer prevention system that can be used at home or in a clinical setting.
This work was funded by the NIHR Invention for Innovation programme (141), Grant reference II-LA-0813-20007.
N. L. Petrova,1,2 A. Whittam,3 A. MacDonald,4 S. Ainarkar,5 A. N. Donaldson,1 J. Bevans,5 J. Allen,4 P. Plassmann,6 B. Kluwe,7 F. Ring,7 L. Rogers,3 R. Simpson,3 G. Machin,3 and M. E. Edmonds,2018, ‘Reliability of a novel thermal imaging system for temperature assessment of healthy feet’, Journal of Foot and Ankle Research, vol. 11, no. 22
Approximately 18 million people in the UK are living with musculoskeletal (MSK) conditions accounting for more than 22% of morbidity. MSK conditions affect the joints, bones and muscles, and also include back and neck pain. In the clinical environment, biomechanical assessment is heavily dependent upon the skills and knowledge of clinicians, which they have gained from years of experience. While there exist standardised methods of recording postural assessments, these methods are cumbersome to use in practice, relying upon clinical intuition to take the required anatomical measurements and interpret the results. To meet this clinical need the research project aims to develop a complete Postural and Functional Assessment System that can be used to perform a wide range of biomechanical assessments in the clinical environment. The proposed system will allow prescription traceability back to assessments, measurements, and clinical decisions, while also enabling clinically defined outcome measures that can track the progress of a patient’s condition and monitor their posture over time. The project has now entered a mature stage where the benefits of the theoretical research could be realised by conducting a clinical validation of the technique. In order to accomplish this validation a large body of experimental work needs to be carried out followed by the data analysis. The results of the experiments need to be compared against the clinical data obtained using standard methods of anthropometric assessment with tactile devices.
Models of player behaviour have applications in computer games ranging from creation of more realistic Non-Player Characters [NPCs] to the detection of ‘Cheat bots’ in multiplayer games. This projects intends to create models of player behaviour which can be applied to these applications. The work also has potential applications in other fields.
Allocation of educators to diverse and rapidly evolving educational programmes of study such as those within Computing and under increasingly tighter budgetary constraints is a non-trivial task given the wide range of subject areas encompassed here, which span information technology, information systems, software engineering and computer science, as well as core areas such as computer forensics and computer games development. Suitability and availability of expertise coupled with an aspiration to limit disruption to existing teaching assignments can often result in first fit solutions that are less than optimal in terms of suitability.
This system is highly sensitive to even small changes, which ripple out through assignments and make it a difficult problem for solution. The ongoing work has resulted in a methodology for profiling modules and, by association, educator expertise that provided a basis for exploring a large number of potential teaching assignments utilising search algorithms. The prototype system limited profiles to what staff had previously delivered. The live system saw the integration of data provided by all academic staff within Computing. This data enhanced suitability profiles by indicating which elements of the curriculum staff would and what they could deliver.
Implemented in a live format for the first time in preparation for the 2013 academic year, the process rapidly achieved a very good solution to this difficult, evidenced by the significantly lower incidence of change requests from colleagues. The solution minimised disruption to existing assignments, highlighted bottleneck areas and distributed 167 teaching units across 44 members of staff while minimising under and over-assignment despite the very large number of unassigned modules left retiring colleagues. Further work includes the inclusion of more elaborate workload balancing heuristics, the application of other algorithms and expansion into other disciplines.
Professor Andrew Ware is working with Aurora International Consulting to develop an AI based system for reviewing method statements and risk assessments for the construction industry. The Government-funded KTP project will enable the University to embed a new capability in the company to use AI technologies. The aim is that the product will be sold to multinationals who can integrate the system into what they do.
The GNU Modula-2 compiler is one of the many language front ends to the GNU Compiler Collection (4.5M lines of code and targetting 58 platforms). The current activities in this area include the completion of whole number overflow detection at runtime and compile time through a compiler plugin. The compiler is operational and work is currently underway to migrate the source code into the GCC git tree. The combination of the Modula-2 front end and GCC results in a modern production quality free software compiler which will be available on many hardware and operating system combinations.
As Director of Research of the Wales Institute of Digital Information, a significant proportion of Professor Andrew Ware's research is cantered on the use of AI modelling techniques to help facilitate the use of health data to help form predictions.
To help internationalise the work that is being done, Andrew has been awarded a Welsh Government grant to visit Mandsaur University in India that have cognate research interests. The visit was meant to take place in 2020 but, due to Covid-19, has been delayed until 2021.
"The Application of Pattern of Life Analysis to the Problem of Early Intrusion and Threat Detection and Classification in Industrial Process Control Systems" undertaken by Peter Donnelly under the supervision of Dr Mabrouka Abuhmida
"The Analysis and Design of a a Socio-technical Business Risk and Vulnerability Assessment Process for Industrial Control Systems" undertaken by Kim Smith under the supervision of Dr Ian Wilson
"Visualisation for Cyber Security" incorporating virtual reality undertaken by Daniel Harris in collaboration with ITSUS Ltd under the supervision of Dr Ian Wilson
Semantically Clustering Medical Referral Letters using Natural Language Processing in collaboration with Digital Health and Care Wales (formerly National Health Service Wales Informatics Service) under the direction of Dr Ian Wilson. The study demonstrates the efficacy of retrieving important features within referral letters written between general practitioners (primary care doctors) and specialist medical consultants to allow for classification into specialisms to help future inform decision making, particularly in situations where access to primary care doctors is limited.
Work into cyber maintainable safety critical information technology and operational technology complex systems undertaken by Kirsty Perrett in collaboration with Thales Ltd under the supervision of Dr Ian Wilson.
Development of a multivariate house price forecasting model using Deep Learning with the aim identifying causal factors affecting the UK housing market and their correlation undertaken by Shaily Jain under the direction of Dr Ian Wilson.
Westgate is a 100K project in the heart of the cybersecurity field. The company approached the university with its algorithm, and the university formed a research team and assignment me the management role of the project. The project main aim is to introduce a level of academic excellence from a trusted third-party organisation with a strong reputation, to help the company to enhance its product, and develop their technology maturity. This academic excellence was essential to gain credibility for their product to get closer to the market. Our objective is to conduct independent validation to build investors and customers’ confidence in the algorithm. One of our main deliverables is an RFC document and produce all the description documentation for the algorithm main procedures and techniques.
This research introduces a new optimisation method for the pre-processing stage of the Point Cloud Library that affords operation on low power embedded devices, such as Raspberry Pi.
Point cloud processing is demanding and previously was not accessible to embedded devices. The Point Cloud Library is used by more than 30 companies around the world including NVidia, Google and Toyota as well as universities and other researchers.
Point cloud processing is also heavily used in autonomous vehicle navigation systems. Elements of this research have been contributed to the library and are also being used by the spinoff company, Thermetrix (podium.care), producing the Podium medical device.
The medical side of the research is still on going to introduce new non-invasive medical devices to help with early detection of diabetes.
The purpose of this work is to determine whether it is possible to use an automated measurement tool to clinically classify clients who are wheelchair users with severe musculoskeletal deformities, replacing the current process which relies upon clinical engineers with advanced knowledge and skills.
Clients’ body shapes were captured using the Cardiff Body Match (CBM) Rig developed by the Rehabilitation Engineering Unit (REU) at Rookwood Hospital in Cardiff.
A bespoke feature extraction algorithm was developed that estimates the position of external landmarks on clients’ pelvises so that useful measurements can be obtained.
The outputs of the feature extraction algorithms were compared to CBM measurements where the positions of the client’s pelvis landmarks were known. The results show that using the extracted features facilitated classification. Qualitative analysis showed that the estimated positions of the landmark points were close enough to their actual positions to be useful to clinicians undertaking clinical assessments.
The work was done in collaboration with Cardiff and Vale University Health Board (CVUHB) and led to several external grants from the NHS and EU.
The output led to clinical trial (IRAS_ID 90356) and was publicised in July 2017 NHS Newsletter sent out to 14,000 employees.
This work is partly supported by the Welsh Government’s ESF-Funded Knowledge Economy Skill Scholarship with the Cardiff and Vale University Health Board’s (CVUHB) Rehabilitation Engineering under Grant MINI 11100. In addition the project received separate funding from the EPSRC first grant scheme (CASE/CAN/06/47]) and CVUHB
Knowledge Economy Skills Scholarships
This work presents a method for non-computationally expensive automatic alignment of cameras that utilises stereoscopic imagery separated at varying distances just below that of the intraocular distance. Here, automatic stereoscopic alignment in real-time is a non-trivial process that relies on calculating the best virtual alignment of camera lenses through image overlaying. This is important as retail 3D camera lenses are not typically sufficiently calibrated for accurate estimates of distance.
The alignment of images allows the filtering of background objects and focuses on points of interest. Imprecision in camera lens calibration leads to problems with the required alignment of images and consequent filtering of background objects.
The algorithm presented in this paper allows virtual calibration within non-calibrated cameras to provide a real-time filtering of images and the consequent identification of points of interest. The proposed method is capable of generating the best alignment setup at a reasonable computational expense in natural environments with partial background occlusion.
Urodynamics is a clinical used to diagnose the pathophysiological reason behind lower urinary tract symptoms with which a patient presents. The test is carried out by taking pressure measurements inside the bladder and rectum and observing how pressure changes during bladder filling and voiding.
The data recorded in urodynamics is usually in the form of a time series containing three pressure traces (the two measured pressures, in the bladder and rectum, and the difference between them which is assumed to be bladder muscle activity). The flow rate of any fluid voided and the amount of fluid that has been pumped into the bladder during the test is also recorded. Occasionally X-ray video is used to aid in diagnosis but in most cases the presence of certain pressure events and key values taken during the bladder voiding cycle are used to make a diagnosis. Here, research into the classification of problems derived from trace data is ongoing.
The pointwise sampling algorithm has been used in adaptive contrast enhancement (ACE) for nearly three decades. To overcome its computational overhead various stepwise sampling algorithms have been proposed, including the HOICE(Highly Overlapped Interpolation Contrast Enhancement) algorithm that addresses the artefacts, such as image distortion and blocking effects, of other stepwise sampling algorithms while keeping the advantage of speed.
However, HOICE’s internal mechanism has not been compared with traditional pointwise algorithms, which may affect the scope of its future application.
This paper establishes a bridge between traditional pointwise sampling and HOICE: it first mathematically analyses the internal mechanism of two typical sampling algorithms (pointwise and HOICE); and then quantitatively demonstrates the similarities between these two kinds of algorithms. Moreover, the results of quantitative measurement suggest that HOICE has an advantage over the pointwise algorithm in reducing the distortion along image edges. Therefore, HOICE has the advantages of both pointwise and other stepwise algorithms making it a reliable sampling algorithm for ACE.
Work completed in collaboration with the Medical Physics Department at Heath Hospital, University of Wales has resulted in a novel technique for increasing the contrast in medical images.
This EPSRC-funded project is a partnership with Newport-based company GXS PDQ Ltd. The company produces business rules for customers in the Retail Fast Moving Consumer Goods and Consumer Electronics sectors, and incorporates these rules into its Product Data Quality (PDQ) service for data quality checking. This enables its customers to identify quality failures in their data. However, the generation of business rules is time consuming, and must be repeated following changes in the business environment, the addition of new products or of new channels for sales, or changes to existing products. The project will involve using artificial intelligence (including semantic and statistical data mining techniques) to automate the generation of these rules.
There is no reliable forecasting service for residential values with current house prices taken as the best indicator of future price movement. This approach has failed to predict the periodic market crises or to produce estimates of long-term sustainable value (a recent European Directive could be leading mortgage lenders towards the use of sustainable valuations in preference to the open market value). Work here, underway in collaboration with colleagues, sees the application of sensitivity analysis to artificial neural networks, trained using multivariate time-series data, which forecasts future trends within the housing market. Prior work in this area has been well received both in academic and media circles, with the changing data that encompasses a sustained period of downward prices offering a further opportunity for research. Work on feature selection and data mining led to a predictive model for residential house price forecasting and has resulted in collaborative work in GIS where the impact of flood plains on house prices will be studied.
We have developed computer models for predicting and mapping crime levels that helped inform the Crime and Disorder Audits of the collaborating Partnerships, in collaboration with local Crime and Disorder Partnerships (South Wales Police, and South Wales Fire & Rescue Service). The work received initial University RS and postdoctoral RA support before EPSRC/CASE funding was secured for two research students.
A system that intelligently interrogates a constantly updated database of crime incidence and provides accurate indicators of where and when crime is likely to be highest would be of great utility in real-time police resource allocation. A limiting factor, however is that crime incidence counts are generally low in relation to crime type, time and space, and subject to randomness. Crime forecast error measures vary inversely with increasing incidence count utilised in estimating time series forecast models. Average crime counts per unit time period and geographic area of at least 25 to 35 are needed before forecast errors become acceptable.
This work details a forecasting framework for short-term, tactical deployment of police resources in which the objective is the identification of areas where the levels of crime are high enough to enable accurate predictive models to be produced. Here, identified hot-spot regions are utilised as the foundation for predictive models.
In collaboration with the Welsh Consultant Haematologists Leukaemia Registry), this work led to a new algorithm for determining salient attributes within cause and effect relationships. Spatial datasets contain information relating to the locations of incidents of a disease or other phenomena. Appropriate analysis of such datasets can reveal information about the distribution of cases of the phenomena. Areas that contain higher than expected incidence of the phenomena, given the background population, are of particular interest. Such clusters of cases may be affected by external factors. By analysing the locations of potential influences, it may be possible to establish whether a cause and effect relationship is present within the dataset.
A collaborative project with Maritime Computer and Technical Services and (the then) P&O Containers resulted in a paradigm for the automated planning of cargo placement on container ships. The methodology was derived by applying principles of combinatorial optimization and, in particular, the Tabu Search metaheuristic. The methodology progressively refines the placement of containers, using the Tabu search concept of neighbourhoods, within the cargo-space of a container ship until each container is specifically allocated to a stowage location. Heuristic rules are built into objective functions for each stage that enable the combinatorial tree to be explored in an intelligent way, resulting in good, if not optimal, solutions for the problem in a reasonable processing time. This body is heavily cited and forms the basis for a resurgence in interest in this area.