[CapoCaccia - announce] Human behaviour & deep learning PhD opportunity

[Pearce, Timothy C. (Dr.)]

Dear Colleagues

Exciting PhD opportunity!

Apply cutting-edge AI to revolutionise human activity recognition using wearable accelerometers. Unlock insights from massive datasets, improve healthcare, and shape the future of personalised interventions.

Join our team BERG https://lnkd.in/g5ZrucBM and LLRG https://lnkd.in/ebXAgB2B !

#AIResearch<https://www.linkedin.com/feed/hashtag/?keywords=airesearch&highlightedUpdateUrns=urn%3Ali%3Aactivity%3A7282344980537348098>
#PhDPosition<https://www.linkedin.com/feed/hashtag/?keywords=phdposition&highlightedUpdateUrns=urn%3Ali%3Aactivity%3A7282344980537348098>
#HealthTech<https://www.linkedin.com/feed/hashtag/?keywords=healthtech&highlightedUpdateUrns=urn%3Ali%3Aactivity%3A7282344980537348098>. Project with Alex Rowlands<https://www.linkedin.com/in/alex-rowlands-0699b7146/> and Tom Yates<https://www.linkedin.com/in/tom-yates-18737b2b1/>.

Project title: Human Activity Recognition using Self-supervised Learning AI Architectures for Wearable Accelerometers in Free-Living Scenarios

Background: Wearable tri-axial accelerometers have been demonstrated as a capable technology for monitoring human activity levels and generating behavioural feedback for preventing chronic disease and comorbidity. Activity levels, sleep and sedentary behaviour are routinely classified from accelerometer data based on signal magnitudes and/or orientation, but deeper insights may be obtained by identifying specific activities and their complex temporal relationships – so-called human activity recognition (HAR). For instance, deeper understanding of daily behaviour profiles across sub-types of type 2 diabetes could support personalisation of interventions and disease management.

While accurately recognising and interpreting human activities from wearable accelerometer time-series data for patients in the community offers potential to unlock valuable insights for improving healthcare and healthy ageing, accurate time-resolved HAR in a free-living context is still challenging. There is a lack of ‘ground truth’ data in free-living contexts and many datasets are collected in artificial laboratory settings with limited wider application to lifestyle research. Moreover, traditional data analysis (e.g. probabilistic and statistical) methods can struggle with the complexity of these data to capture the variability and the rich repertoire of human behaviour in the free-living context. Hence there has been a call to apply recent advances in AI to tackle large unlabelled free-living accelerometery datasets to unlock behavioural insights.

This project will explore the development of self-supervised AI architectures to unlock behavioural insights in large accelerometry data-sets. These architectures promise a reduced need for manual labelling, improved generalisation, performance gains on downstream tasks, capture of complex temporal dependencies, adaptability to real-world variability and the potential for personalisation and scalability.

Please contact me to discuss t.c.pearce at le.ac.uk<mailto:t.c.pearce at le.ac.uk>
Application link and more project details here: https://lnkd.in/egQrqXqg


Best, Tim.

--

Tim C. Pearce, PhD FInstP SFHEA (he/him/his)

Reader in Bioengineering


Advance notice of leave: away from 9th December onwards until New Year.



Biomedical Engineering Research Group (BERG)
School of Engineering
University of Leicester | University Road | Leicester | LE1 7RH | UK

t: +44 (0)116 223 1290
e: t.c.pearce at le.ac.uk

w: https://le.ac.uk/people/tim-pearce

w: https://le.ac.uk/engineering/research/berg
li: https://www.linkedin.com/in/tim-pearce-808aaa83/


shared calendar: here<https://outlook.office365.com/owa/calendar/97def62d5929470cad007b8b5ba0b354@leicester.ac.uk/49a25705a2c448018ebdf65992a4126c6579692957646256408/calendar.html>


[cid:dc861d4f-f7f3-4dd8-87a1-00ecff8fb3d9]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mail.capocaccia.cc/pipermail/announce/attachments/20250126/9164f87c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-ebhxygzb.png
Type: image/png
Size: 7548 bytes
Desc: Outlook-ebhxygzb.png
URL: <http://lists.mail.capocaccia.cc/pipermail/announce/attachments/20250126/9164f87c/attachment-0001.png>