The first signs of mental decline could be recognised by computer to enable people to refer themselves for medical care and support.
Currently, only half of people with dementia ever receive a diagnosis so researchers are trying to find new ways to persuade people to seek help. Lancaster University is leading a project to see if computer interaction can offer new opportunities for self-referral.
The £700,000 Software Architecture for Mental Health Self management (SAMS) project will look at the use of data and text-mining techniques, combined with adaptive user interfaces to detect early signs of cognitive decline from the way people use their computers.
Professor Pete Sawyer from the School of Computing and Communications
said: "SAMS aims to contribute by exploiting the fact that the elderly increasingly use email and electronic social networks to keep in touch with friends and family.
“These electronic media provide a way for us to develop techniques to identify significant changes in the way people communicate and how they interact with the computer or mobile device. This is a really hard challenge for computer science but there is also the further challenge of how to use the results in a way that increases the rate of self-referral without causing unnecessary distress or alarm.”
The SAMS project is funded under the EPSRC Working Together call and is led from the School of Computing and Communications by Professor Pete Sawyer, Dr Paul Rayson and Professor Alistair Sutcliffe, and is joint with Manchester University and Kings College London. The project is supported by the Dementias Neurodegen Network (DeNDRoN), The Alzheimer's Society, Microsoft Research, the University of British Columbia and Johns Hopkins University School of Medicine.