Skip to menu | Skip to content |

Centre for Health Informatics

Completed research projects within the Centre for Health Informatics

This page gives details of all research activity now completed within the Centre for Health Informatics, divided by Research Group :

eHealth and Telecare
High Dependency
Public Health

eHealth and Telecare

The increased use of ICT in the organisation and delivery of healthcare offers the potential to increase the information and knowledge available to the clinician to support their decision-making. In this way the development of decision support systems forms a key component of the research programme of the Centre. Related to this is research seeking to automate the process of acquiring medical knowledge from the web, and developing tools to manage such knowledge more effectively.

However, as well as supporting the clinician, there are also opportunities to help the patient better manage their disease. This is particularly the case in relation to chronic diseases such as diabetes and hypertension, where information systems are being developed with the objective of making the patient better informed about their disease condition.


REALITY: an Evaluation Study in Home Telecare
Investigators: Abdul Roudsari, Howard Leicester, Fiona Harvey, Roland Petchey, Peter Levine, Derek Cramp , Ewart Carson and other members of a European consortium
Funding:TBC
Funding source:TBC
Duration:TBC
Overview: This EU-funded project focuses on a range of evaluation aspects associated with home telecare. The project consortium, led by CHI, includes clinical partners from Estonia, Portugal and the UK, together with a UK technical partner responsible for the provision of hand-held computers being used by patients in their home settings. Patients across a range of urban and rural settings in these countries are collecting both clinical and quality of life data associated with the management of their chronic disease. The chronic diseases involved are chronic respiratory failure, bronchial asthma, chronic heart failure, hypertension and type 2 diabetes.

Socio-economic evaluation is now being undertaken from the perspectives of the patient, the healthcare professional and the health manager. Preliminary findings indicate that issues needing to be addressed in order to achieve a successful telecare operation include: proper patient selection methodology; the need for suitable and adequate training for all those involved in the delivery of care process; improved co-operation and support of telecommunication operators; the key role of nurses and their involvement; the need for clear user instructions in simple language; and simplification of data input, including less complex question sets.


Towards The Semantic Web: The Automation of Knowledge Acquisition from the Medical Web
Investigators: Mohammad Ali Eljinini, Nazar Sarhan (Al-Isra University, Amman) and Ewart Carson
Funding:TBC
Funding source:TBC
Duration:TBC
Overview: The current Web contains a wealth of information in the form of natural text. HTML documents are designed for human consumption and not for machines. Adding machine understandable representation to web documents will enable contents-guided searching and reasoning. This will lead the web to its full potential. It is impractical to go back and annotate the current web manually. Such process is labour intensive, prone to errors, and requires expertise with new complex technologies. The aim of this research is the development of techniques for extracting knowledge from medical-related websites with the guide of ontologies. The first part of this work is about the ontology development and the possibility of reusing exiting ontologies. The second part concentrates on methodologies for extracting the right information which is to be transformed into machine understandable form.


Providing Patient Information for Managing Chronic Disease
Investigators: Emma-Jane Berridge, Mira Vogel, Abdul Roudsari and Ewart Carson
Funding:TBC
Funding source:TBC
Duration:TBC
Overview: ICT-based systems are being designed and evaluated so as to enable patients to obtain information at home about the nature of their chronic disease and how they should better manage it. The activity has initially focused on the provision of information for individual diseases, though the aim is to produce an integrated, web-based, CAL system providing information about diabetes, hypertension and asthma, which are amongst the most common chronic diseases. By using appropriate techniques of knowledge management and adaptive learning it is increasingly possible to tailor the information provided to the needs of the specific user. This application domain in health informatics has great growth potential consequent upon the increasing empowerment of patients as customers in relation to healthcare provision. By adopting sound methodology for system design and evaluation, web-based information packages are being produced that meet the needs of the patient as the system user.


Evaluating the Impact of Digital Libraries on Knowledge about Healthcare
Investigators: Gemma Madle, Patty Kostokova and Abdul Roudsari
Funding:TBC
Funding source:TBC
Duration:TBC
Overview: Recent years have seen an explosion in the availability of healthcare information for both patient and healthcare professional. Are patients applying the information in an appropriate manner, which will assist them and their clinician in managing their health? Or is the information retrieved by patients inaccurate, or simply misunderstood? For healthcare professionals this availability should enhance evidence-based decision-making. However, recent research suggests that senior clinicians can be reluctant to allow more junior staff access to current evidence-based information, preferring instead to provide it for them or withholding it altogether. This provides further support for the importance of medical digital libraries that are accessible to all via the Internet and their potential role in changing the culture of the health working environment and the relationships within it.

The impact of such ehealth technologies on the knowledge and attitudes of health professionals and patients is being investigated. Approaches include recruiting panels of health professionals and patients in a longitudinal study to investigate the impact of the National electronic Library for Communicable Disease on professional practice and patient behaviour.


High Dependency

The intensive care unit, operating theatre and other high dependency hospital environments are increasingly dependent on advanced healthcare technologies including applications of ICT. These are particularly data-rich environments and hence a major challenge is to extract the maximum of information to support clinical decision-making.

The development of tools to support the interpretation of such data is a key focus of the research of the Centre. This includes developing advanced mathematical techniques to detect trends in heart rate and to interpret EEG signals which, for example, may provide early warning of possible brain dysfunction.

Modelling also forms an important component of the research programme. At one level, modelling can provide added insight into the changes of blood flow that occur following cardiac surgery. At another level, modelling can be used as an aid to planning the management of interacting high dependency units, predicting length of stay of the acutely ill patient as they progress from operating theatre to intensive care unit to step-down unit etc.


Wearable Computers in the Operating Room
Principal investigator: Peter Weller, Qi Ma, Helen Petrie (Centre for HCI Design), Gerlinde Mandersloot (Royal London Hospital), Darren Morrow (Royal Adelaide Hospital, Australia), Arjuna Weerasinghe (Harefield Hospital)
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The modern operating theatre is an environment where high technology is commonplace. However, whilst technology can greatly improve patient care, the very presence of equipment can result in some basic problems such as being able to efficiently observe and comprehend patient specific data. Often the key personnel in the operating room, such as the anaesthetist and the surgeon, have to constantly divert their attention between the patient and the monitoring system. This situation cannot be optimal for patient care.

The aim of this project is to develop and evaluate an improved patient monitoring system based on wearable computer technology. It is envisaged that the key clinicians will each carry one of these paperback size computers. These will process the data collected by the patient monitor and project selected variables onto the clinician's normal view. With this approach both the patient and key variables can be observed concurrently.

Chaos in the EEG during Cardiac Surgery
Principal investigator: Peter Weller, Jacques LeFevre (EC-Lille), Marc Baguelin (Cambridge), Richard Pottinger (St. Bartholomew's Hospital)
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The electroencephalogram (EEG) is a measure of the electrical activity of the brain. It can reflect the state of a patientÆs brain and their level of consciousness in a characteristic manner. Although primarily used in the diagnosis and management of epilepsy and sleep disorders, the EEG has proved useful in the operating room.

This project uses techniques from the mathematics of chaos and fractals to develop a suite of tools for analysing the EEG during cardiac surgery. Data collected from a previous study are being analysed to provide clinicians with additional insight into the patientÆs condition. Some of the results achieved to date show that a change in some of these variables is associated with clinical procedures and so may give warning of pending trauma.


Modelling of Blood Flow in Cardiac Bypass Surgery
Principal investigator: Peter Weller, Arjuna Weerasinghe (Harefield Hospital), Joe Iannelli (Biomedical Engineering Centre)
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: Cardiac bypass surgery is a procedure by which a blocked cardiac blood vessel is bypassed using a length of vein harvested from the patient. The common method of attaching this graft is by suturing (sewing) the new vein either side of the occluded vessel. This changes the blood haemodynamics and could result in a build up of thrombus so hastening early graft failure.

The work in this study involves is to developing a mathematical model of the blood flow around a single graft. The initial work has only considered a simplified system, but the model will be continually refined. The results are being verified in laboratory tests. The ultimate aim of this work is the development of a decision support tool for surgeons.

Interpreting Data in Complex Environments
Principal investigator: Peter Weller, Llew D'Souza, Alex Thompson (HMS Sultan)
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The ability to diagnose the current condition of a complex system and then predict its future behaviour with that diagnosis is a powerful aide for the monitoring of such systems. Examples of these include high dependency patient monitoring, power generation, aircraft flight control and financial markets.

In this project a multi-layer monitoring system has been developed. Each layer provides the observer with a more detailed insight into the status of the system. The diagnosis is performed using artificial neural networks. The lowest layer consists of a mathematical model of the system. This is configured using the results from the upper levels and provides a snapshot of the future behaviour of the system. A basic prototype has been developed and tested, with further evaluation on-going.


Information Systems to Support Healthcare Resource Allocation
Principal investigator: Mike Hughes, Cliff Morgan and Ewart Carson
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: This project focuses on the potential improvement in cost-effective healthcare delivery by the adoption of a computer-assisted patient scheduling system (CAPSS). This involves showing that providing healthcare managers with more complete and accurate information about the projected availability and demand for healthcare resources improves the ability to control the operational performance of the healthcare system. The ability to deliver this necessary information to the control system in a timely and efficient manner is only realistically attainable through the computerisation of the patient scheduling system, and hence the deployment of CAPSS. CAPSS is being designed using a number of innovative techniques. These include modelling patient physiology, modelling the sequence of admissions of individual patients through the hospital environment (including the operating theatre, intensive care unit and ward), and the resultant effects of this on projected resource requirements and quality of care protocols.

Detecting Local Trend Changes in Heart Rate
Principal investigator: Peter Weller and Mark Ansermino (University of British Columbia)
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The detection of local trend changes in heart rate is important for patient care and in assessing the quality of physiological records. However, the methods needed to achieve this are often non-trivial. In this work the Holt-Winters method is being used to predict future values of heart rate. These are then compared with the actual recordings. An algorithm is then used to identify change in the underlying process. The method has been evaluated on recorded patient data and shown to successfully identify clinically-subtle changes. This evaluation is ongoing with new patient data. Additionally, an on-line version is being developed for clinical trials.


Public Health

Changes in the organisation of healthcare delivery mean that public health professionals are more widely dispersed across healthcare organisations. Given this situation, ICT becomes increasingly important in ensuring effective communication amongst the public health community. Assisting in this process is one of the aims of the research programme being undertaken in the Centre.

Modelling the prevalence of disease helps in assessing the need for health care facilities. When coupled with cost modelling, information can be obtained to support decision making in relation to changes in service provision; one example is the provision of screening services to identify retinopathy in the diabetic patient.

Modelling, making use of soft systems methodology, can yield insights regarding the real needs for specific healthcare services, and offers powerful tools which can, for example, assess the worth of National Service Frameworks, for instance that which has been developed in relation to services for the diabetic patient. Such modelling provides a powerful aid to the health policy maker.

Other research focuses on the needs of those with impaired vision, including the role of electronic systems in helping to ensure that the patient is better informed.


A Soft Systems Methodology Analysis of the NSF for Diabetes
Principal investigator: Kanwal Karim, Derek Cramp and Ewart Carson
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The implementation of new policy all too often entails the resolution of complex, interacting requirements. Such is the case with the National Service Frameworks (NSFs). In this project an approach framework has been developed that can support policy analysis and help define the subsequent process of its implementation. This has involved an analysis of the National Service Framework for diabetes using Soft Systems Methodology (SSM).

SSM was found to be of value in solving unstructured and poorly delineated problems where goals are either not properly defined or are debatable. Using the visualisation, conceptualisation and layered thinking tools of SSM, desirable and culturally feasible changes became apparent. It is shown that, using SSM, effective communication is the prime requirement for effective implementation of the NSF for diabetes. This relates both to communication within individual organisations involved in the delivery of diabetic care as well as between organisations. Getting this aspect right would help drive the other important areas which are availability of services, access to information and resource allocation.

In order to fulfil the vision in these three areas, careful consideration needs to be paid towards organisational, cultural, technical and economical barriers. These do seem like basic requirements, but the power of ideas embodied in SSM allows these core issues to be adequately addressed. Analyses such as that undertaken in this studydescribed in this paper become all the more timely with the move towards a health service which is increasingly underpinned by an infrastructure based on information and communications technologies.

Knowledge Management from a Public Health perspective: a TB Case Study
Principal investigator: Anjana Roy, Patty Kostkova, Anand Kumar (Leipzig University), Julius Weinberg, Gemma Madle and Ewart Carson
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: Formal ontological representation of clinical conditions and disease management is a key methodology in helping to ensure that the complex knowledge of disease treatment, control and prevention can be represented, stored and accessed in the most appropriate way to help medical professionals in their decision making. This is of particular importance for the public health domain where the concern is about the affect of the disease on populations rather than individuals.

The existing evidence-based knowledge can best be used by professionals if incorporated into care pathways (formal or informal) which relate to the sequence of actions necessary for accurate management of diseases. Hence there is the need for an ontological framework to be built around care pathways in order to allow the professionals to access the most relevant information at the time of making a decision.

This project is exploring these issues in the context of tuberculosis care pathways. A formal ontological representation has been developed, designed to serve the needs of information retrieval associated with this particular disease.


Health Informatics and the Eye Network in North Central London SHA
Principal investigator: Howard Leicester
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: The Department of Health has proposed new care pathways to improve eye care and social services in local areas. The initiative complements the VISION 2020 programme from the World Health Authority (WHO) which aims to reduce avoidable sight loss by 2020.

North Central London Strategic Health Authority (SHA) responded in 2004 with an "eye network" of planners, clinicians, patients, analysts and commissioners to oversee progress across the five local PCTs. Clinical subgroups have been established. A complementary group covering epidemiology and information for planners and patients has also been formed at Dr Leicester's request. The information group is now co-chaired by an ophthalmic epidemiologist and North Central London's Director of Public Health.

The main causes of UK sight loss are age-related. Commissioners have therefore been reminded of overlaps with the Single Assessment Process for older people (SAP) and recent Government interest in long-term conditions. Prevalence modelling has been proposed, based on North London studies and an equivalent NHS approach in diabetes. The limits of hospital activity data, as a source of monitoring information, have been identified.

In contrast, there are immediate opportunities to develop electronic systems in patient information provision. It is technically feasible to convert electronic material into different output formats for use away from computers by people with special communication needs. This principle of "transcoding" was introduced in version 3 of the eGovernment Interoperability Framework (eGIF) but details and working examples were never given. The Department of Health might now give this serious attention along with its new eye care pathways.


The Prevalence of Diabetes, Screening for Diabetic Retinopathy and its Cost Implications
Principal investigator: Derek Cramp, Fiona Harvey, Malgosia Wilinska, Abdul Roudsari and Ewart Carson
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: This study was commissioned by the North Central London Strategic Health Authority (NCLSHA), and funded by the Camden Primary Care Trust on behalf of Camden and Islington PCTs, to help inform them of the resource and service delivery implications arising from implementation of the National Service Framework for Diabetes and to provide a background for those who have to formulate policy and make decisions concerning the provision of diabetic retinopathy services in the context of the National Service Framework for Diabetes. The study focused on three specific concerns within the NSF standards, namely: the detection of undiagnosed diabetes, screening for diabetic retinopathy and diabetic care in primary care. Knowing what proportion of the local population served by the five Primary Care Trusts in the NCLSHA has diabetes and an estimate of the number undiagnosed and who represent a potential burden on services is crucial. A prevalence model, based on epidemiological and demographic data, was developed. The information generated by the model together with the results of a survey of current eye screening services in the area provided information as to how new to address the problem of assessing potential demand and costs of systematic retinopathy screening services.

a Bayesian modelling approach which has been shown to yield more reliable estimates of parameters which characterise insulin sensitivity and glucose effectiveness in patients with type 2 diabetes. Moreover, this model has been shown to yield insight into the progression of the disease in the two years following its diagnosis, in terms of these model parameters and their relationships with clinical measures of glucose control.


Insights Achieved Using Systems Methodologies
Principal investigator: Daragh Fahey, Kanwal Kalim, Derek Cramp, JA Muir Gray and Ewart Carson
Funding: TBC
Funding source: TBC
Duration: TBC
Overview: Systems science and systems thinking are central to much of the activity carried out within CHI. In particular soft systems methodology is being widely applied as a means of gaining insight into the nature of complex organisations such as occur widely across the NHS. This methodology is particularly appropriate for dealing with ôhuman activityö systems. It has been successfully adopted in defining the need for networks by means of which the now dispersed public health professionals will be able to communicate effectively and in the development of appropriate designs for such public health networks (see list of publications). Another application relates to the provision of services for diabetic patients.