Background

The aim of this work is to build open source technologies, including apps, that are able to assess cognition and behaviour of relevance to conditions in Brain Health. It will focus on some of the most important conditions facing society today, like Alzheimer’s Disease, which is estimated to cost the UK £26 billion per annum. Digital phenotyping methods have a huge potential to track cognitive trajectories remotely, longitudinally, continuously, precisely, economically and at population scale. This makes them especially appropriate for measurement at the prodromal stage. They have widespread applicability, for example, in the development of more effective clinical trials, but also as the building blocks for precision public health. Recently, a number of citizen science projects have emerged which make digital phenotyping methods available to large, open, populations of participants who never attend traditional in-clinic visits. These have huge potential, but also introduce many challenges. This project will investigate, through practical studies, the utility of this new digital method of research.

Research Experience, Research Methods and Training

The work will entail developing technologies, either active or passive, to detect subtle longitudinal changes in behavior and cognition. It will involve designing apps that can be used to engage large, remote populations, and evaluating their efficacy. It will be necessary to engineer solutions that can be deployed at significant scale. It will also involve designing studies that produce statistically robust outcomes, working with users and ethics committees to validate the approach, and analyzing the collected data so as to maximize clinical insight. The research will need to tackle challenges from diverse and emerging fields including: digital phenotyping, informatics, statistics, gamification, cognitive psychology, passive measurement, and citizen science.

Prospective Candidate

Candidates will be required to hold, or expect to gain, at least an upper second class honors degree or Master’s degree in computer science or a related discipline. Candidates must be able to demonstrate an outstanding aptitude for software engineering, a good understanding of statistics and epidemiology, and a strong interest in the design or evaluation of digital health technology. Experience with applied machine learning techniques, or cognitive psychology will be helpful.

 

Find out more

Supervisors

Chris Hinds

Senior Research Fellow

Chris Hinds

Martin Landray

Professor of Medicine and Epidemiology; Deputy Director, Big Data Institute

Martin

John Gallacher

Professor of Cognitive Health, Director, MRC Dementias Platform UK

John Gallacher