Published on: 17th May 2018
Over 35 researchers enhanced their knowledge of Natural Language Processing (NLP) at a recent workshop, organised by The Farr Institute and Health Data Research UK (HDR-UK) Wales and Northern Ireland at Swansea University.
The secondary use of Electronic Health Records (EHRs) has transformative potential for how health care research is conducted, opening up new possibilities in areas of business intelligence, observational research, clinical trial recruitment and decision support. However, as much as 80% of the data in EHRs are known to be locked in the form of unstructured text, making this information ‘invisible’ for standard analysis techniques. NLP software, such as GATE, provide a method to make sense of this unstructured text and transform it into terminology which can then be easily utilised by researchers and the NHS.
The two day NLP workshop, which was delivered by Dr Angus Roberts and Dr Sumithra Velupillai from King’s College London, focused on providing an introduction to the field of clinical NLP, from its origins before the advent of ‘Big Data’, to the current state of the art, comprising of information extraction algorithms processing millions of documents on large compute clusters and clouds. In addition, the course, via a series of talks and practical sessions, provided an appreciation of the complexity that different NLP problems pose.
Dr Angus Roberts, said, “The wide variety of participant backgrounds was very positive, and made for stimulating discussions both in and out of the classroom. It was exciting to see the growing interest in NLP of clinical text in Swansea and more widely in Wales. I’m looking forward to seeing what NLP gets developed now.”
A wide range of organisations were represented at the workshop including: Dementias Platform UK, Diabetes Research Unit Cymru, Aneurin Bevan UHB, Abertawe Bro Morgannwg UHB Cardiff & Vale UHB, Cardiff University, Swansea University, HDR-UK, ADRC-Wales, SAIL Databank, NHS Wales Informatics Service, NHS Digital and The National Centre for Population Health and Wellbeing Research.
Shang-Ming Zhou, Senior Lecturer, Farr Swansea University, commented, “This training course was very worthwhile, with clear structure and explanations. The hand-on-exercises provide excellent examples using GATE to analyse free texts.”
Dave Price, Improvement and Innovation Product Specialist, at NHS Wales Informatics Service, said, “I found the course, hosted at Swansea University by Angus Roberts and Sumithra Velupillai, enlightening. It gave real-world solutions in the NLP domain of medical free text.”
Ashley Akbari, Senior Research Manager & Data Scientist of The Farr Institute Swansea University & HDR-UK Wales and Northern Ireland, added, “As a centre, we are always looking to be involved in and collaborate on exciting and innovate research, which increases the impact for both academics and the NHS in Wales and the UK. Part of this is ensuring that our researchers, and those we collaborate with, are always learning and developing new skills from leaders in the field, and we were very pleased to organise this opportunity with Dr Roberts as part of our ongoing collaboration.”
Enquiries to Sarah Toomey, Communications Officer, Farr Institute Swansea University, firstname.lastname@example.org