Our Partnership

University of South Wales

With more than 30,000 students from 120 countries, the University of South Wales (USW) Group is a major player in UK higher education.

Within the United Kingdom, USW is unique in the breadth of its role, encompassing a modern university and two subsidiaries in Wales’s national conservatoire, the Royal Welsh College of Music & Drama and The College Merthyr Tydfil.

USW is a powerhouse in applied research used to shape major decisions. As a major public policy think-tank, it offers independent advice to government, industry and employers across the UK on health, education, economic growth, social policy and governance.

USW has a long and proud track record of both teaching and research in computer science. In recent years, much of this has been based on work carried out in collaboration with business and industry in the application of Artificial Intelligence and Data Science to help solve real-world problems.

For example, recent research has resulted in algorithms and techniques that can be used to cost-effectively diagnose cancer in countries, such as Uganda, which was the primary location for the testing, with developing economies that cannot afford solutions that are available in more affluent areas.

More locally, work in collaboration with NWIS has used data science to builds models that can predict when someone is likely not to attend a scheduled appointment with a clinician. Such ‘did not attend’ scenarios are costly in terms of time and money. Knowing who is likely to miss an appointment can help inform ameliorating interventions to reduce such occurrences.

Another recent project has centred around the use of an Artificial Intelligence technique known as Natural Language Processing to enable handwritten and free-text medical records to be processed, coded, and analysed automatically.

Natural Language Processing has the potential to transform the use of medical records from static, hard to navigate documents, into data that can automatically be used to chart individual patient progress and, when combined with potentially millions of other documents, help facilitate the detection of patterns and trends.