Digital Biomarkers for Mobility Decline and Falls in Dementia

About the project

Can wearable technology detect the earliest signs of mobility decline before they become visible in the clinic?

Falls and mobility problems are among the most disabling consequences of dementia, affecting independence, quality of life, and healthcare utilisation. Yet current clinical assessments often identify changes only after significant decline has occurred. This PhD will investigate whether continuous real-world movement data captured using advanced wearable sensor technology can reveal subtle signs of mobility decline long before conventional assessments detect problems. By combining cutting-edge wearable devices, digital biomarker development, and artificial intelligence, the project aims to transform how mobility decline and fall risk are detected, monitored, and ultimately prevented in people living with dementia.

Working at the intersection of clinical neuroscience, biomedical engineering, and data science, you will analyse high-resolution movement data collected from people across the spectrum of cognitive ageing, including individuals at risk of dementia and those living with established disease. You will explore how patterns of gait, balance, activity, and everyday movement can be used to identify early changes, predict future decline, and improve personalised approaches to fall prevention.

The project offers a unique opportunity to work with state-of-the-art wearable technology, real-world movement datasets, and clinical cohorts. You will receive training in: • wearable sensor technologies and digital health • signal processing and time-series analytics • machine learning and predictive modelling • digital biomarker development • clinical and cognitive assessment • translational research and healthcare innovation

Your findings could contribute to the development of new digital tools that allow clinicians to identify mobility problems earlier, stratify fall risk more accurately, and support people with dementia to remain active and independent for longer.

We are seeking highly motivated candidates with backgrounds in Engineering, Computer Science, Artificial Intelligence, Data Science, Mathematics, Neuroscience, Biomedical Science, Physics, or related disciplines. Applicants should have strong quantitative and analytical skills, an interest in interdisciplinary research, and enthusiasm for applying technology to address major healthcare challenges.

This studentship provides an excellent opportunity to develop expertise in one of the fastest-growing areas of healthcare innovation: the use of wearable technologies and digital biomarkers to improve outcomes for people living with dementia and other age-related conditions.

Professor Michael Hornberger

• Personalised cognition in preclinical and clinical dementia
• Cortical and subcortical neuroimaging changes in dementia
• Sensors/Wearables to detect real-world changes in dementia

Professor Liudi Jiang

• Wearable sensors
• Healthcare technologies
• Musculoskeletal biomechanics

Entry requirements

1st or 2:1 degree in a relevant discipline and/or second degree with a related Masters.

Fees and funding

Due to funding restrictions this position is only open to UK applicants. This PhD studentship is fully funded for 3 years by Medical Innovation Fund of the University of Southampton. Amount of stipend plus fees at UK residency rate only

How to apply

Please complete the University's online application form, which you can find at https://www.findaphd.com/phds/project/digital-biomarkers-for-mobility-decline-and-falls-in-dementia/?p199828

You should enter Prof Michael Hornberger as your proposed supervisor. To support your application provide an academic CV (including contact details of two referees), official academic transcripts and a personal statement (outlining your suitability for the studentship, what you hope to achieve from the PhD and your research experience to date).

Contact us

Informal enquiries relating to the project or candidate suitability should be directed to Prof Michael Hornberger ( m.hornberger@soton.ac.uk ) and Prof Liudi Jiang ( L.Jiang@soton.ac.uk ).

Research areas