Characterising Cognitive Decline
Research type
Research Study
Full title
Characterising Cognitive Decline in Ageing and Dementia using Natural Language Processing
IRAS ID
212014
Contact name
Peter Garrard
Contact email
Sponsor organisation
St George’s University of London and St George’s University Hospitals NHS Foundation Trust
Duration of Study in the UK
1 years, 10 months, 29 days
Research summary
In the early stages of Alzheimer’s Disease (AD) problems with day-to-day memory are often accompanied by milder difficulties with other areas of cognition, such as planning ahead, visual processing skills and language ability. All problems become more prominent as the disease progresses. The analysis of language change has led to particularly revealing insights, which may aid in the prediction of onset and prognosis of AD. Previous studies have found that analysing spoken language using a range of computational methods called Natural Language Processing (NLP) can distinguish individuals with AD from those without, very early on in the disease. Furthermore, analysis of writing can reveal changes in language apparent before other symptoms appear. We wish to build on this work, by investigating spoken discourse in patients recently diagnosed with probable AD, compared to cognitively intact ageing, over one-year. The aim is to identify components of speech, or ‘linguistic factors’, which are sensitive to brain changes in AD. We will investigate novel techniques for collecting speech samples, in order to improve current methods used clinically and in research. Furthermore, we will investigate written language over one year, to analyse which linguistic factors are sensitive to change, and look at MRI scans to see where changes in the brain may relate to changes in language.
REC name
London - Dulwich Research Ethics Committee
REC reference
16/LO/1990
Date of REC Opinion
25 Nov 2016
REC opinion
Further Information Favourable Opinion