| 2003 |
|
YEAR BOOK |
University of Ulster
|
Research within the Faculty of Informatics
|
|
��Artificial Intelligence
��Information Engineering ��Intelligent Systems ��Software Engineering ��Medical Informatics. |
The core research groups have grown strongly within the University and Faculty selective research strategies, and full-time research staff and post-graduate research students have increased from 17 in 1986/87 to over 70 at present. Much of the research work of the Faculty involves industrial and European partnerships, in the key areas of distributed database systems, software engineering, knowledge-based and expert systems, system design methodologies, human-computer interaction/vision systems, object-oriented programming consensus, statistical modelling and mathematical education. The Faculty of Informatics has strong research links with universities and research institutions in USA and China, and with European, American and Japanese companies.
Knowledge and Technology Transfer
The research output is strategically aligned to the benefit of the regional economy by stimulating and supporting knowledge and technology transfer. The Northern Ireland Knowledge Engineering Laboratory (NIKEL) and the Centre for Medical Informatics provide outlets for the application and exploitation of the core research. A number of spin-out companies and joint ventures have also been formed in recent years. The establishment of the Centre for Software Process Technologies (CSPT), funded by Invest Northern Ireland, is a recent major development in the Faculty within the Software Engineering RRG. The CSPT was established on 1st September 2002 and is engaged in developing standards and processes for improving software quality.
European Partnerships
In recent years the European Commission has strongly supported the research work undertaken by the Faculty through substantial funding of a number of major research projects. Currently the Faculty is involved in six large collaborative projects with European partners from academic, commercial and government institutions:
|
��COSMOS - cluster of systems of metadata for official statistics
��ICONS - intelligent content management system ��MEDICATE - the control, identification and delivery of prescribed medication ��METANET - a network of excellence for harmonising and synthesising the development of statistical metadata ��QUDOS - quantum tunnelling device technology on silicon ��SENSEMAKER - a multi-sensory, task specific, adaptable perception system. |
The Faculty is also actively engaged in a number of EU networks of excellence in the areas of machine learning, knowledge discovery, data mining, and the use of artificial intelligence in dynamic and uncertain situations: MLNET, KD Net, EUNITE, and ELSNET.
Conferences and Workshops
Members of the Faculty of Informatics are active in organising national and international research workshops and conferences. In 2002, the Faculty hosted the First International Soft-Ware Conference on Computing in an Imperfect World and the Conference on Applied Statistics in Ireland; in September 2003 the Faculty hosted the Sixth Irish Machine Vision and Image Processing Conference.
The Medical Informatics RRG
The Medical Informatics Recognised Research Group applies computational techniques to assist medical research. The work involves collaboration with local hospitals, other university departments, technology providers, local health charities, and patients. The main activities comprise research in the areas of knowledge acquisition and dissemination, telemedicine, and medical imaging.
Knowledge Acquisition and Dissemination
Techniques can be applied to extract �knowledge� from clinical data. This can provide a better understanding of medical conditions and can be used to provide decision support.
|
��Research in
audiology
attempts to answer the question:
Can we assist the audiologist in quantifying hearing deficit by analysing evoked potentials?
A stimulus is applied to the patient, in this case a sequence of acoustic clicks. Electrical data is recorded from the patient�s scalp using electrodes. Artificial intelligence techniques are used to inspect the waveforms and determine normal or abnormal hearing. A typical project involves pre-processing to clean the data, followed by feature extraction and classification, often using mathematical modelling and/or data mining algorithms. If an abnormality is detected, this is then used to alert the clinician and may assist a diagnostic decision.
��Other areas under investigation include: �� Cardiology: Which recording positions are most important for detecting malfunction from the electrocardiogram? � Diabetes: Can we use a data warehouse to deduce an optimal therapeutic plan for a patient, given his/her symptoms? |
Telemedicine
Members of the group lead the European Union MEDICATE project in the management of drug medication over an intranet. Other areas of interest include �remote monitoring of ECG� and �decision support using the personal data assistant�.
The close relationship with local hospitals is exemplified by the �state-of-the-art� facilities offered within the Ulster Institute of Telemedicine. Facilities include a technology development laboratory, an E-learning education suite, and a lecture theatre with leading edge multimedia equipment and video conferencing infrastructure. Multimedia has been used to develop a pioneering training package for people newly diagnosed with multiple sclerosis. Other areas of interest include training for people with diabetes, chronic obstructive pulmonary disorder and heart disease.
Research has developed revolutionary techniques to help stroke victims regain use of their upper limbs, with the help of virtual reality. This involves the patient wearing a head-mounted display, which provides a sense of immersion into a virtual world, and a flexible glove connected to position and orientation sensors. These enable the patient's hand and arm movements to be tracked in the virtual environment, providing visual feedback to the patient. The system can be configured to exaggerate small movements, increasing the feeling of achievement and improving patient motivation.
Medical Imaging
Image processing techniques have been developed to assist with �ultrasound foetal monitoring� in Obstetrics and �automated detection of lesions in the retina associated with age-related macular disease� in Ophthalmology.
Contact: Professor Sally McClean, Director of Research, Faculty of Informatics, University of Ulster; Email: [email protected]