Hard to concentrate on anything only poor Aisling Murphy this morning. Wrote this over a year ago. Not one man in Ireland will think twice about his safety if he pulls on his runners today. Plenty of women will. We won't fix it until we acknowledge it. irishtimes.com/sport/other-sp…

Here Behavioural Science & Health UCL we are looking for adults aged 50+ to take part in a research study exploring the influence of physical activity 👟 and sleep 😴 on #dementia risk using wrist-worn monitors ⌚️ 💵 £20 gift voucher for taking part! Interested? DM me for more info

Our new preprint is now available online 📢 (!!!)

People are seeing into themselves in ways not possible before and are finding new ways to act on what they learn. The effect on health and lifestyle is likely to be profound econ.st/38NqEJm

Our new pre-print is now online! We use self-supervised learning for human activity recognition using 700,000 person-days of data in the UK Biobank. Fantastic collab between Hang Yuan, Shing Chan, Catherine Tong, David Clifton, Aiden Doherty ! 📄arxiv.org/abs/2206.02909 👇

"In the 1960/70s, less than 1% of students depicted scientists as female. But the percentage of women in the "draw a scientist" sketches has increased over time, reaching an estimated 34% by 2016." science.org/content/articl…

A preprint of our new paper is now available online! 🥳 "Digital health technologies (📱+⌚) and machine learning augment patient reported outcomes (📄) to remotely characterise rheumatoid arthritis." I summarise our main findings in the blog post below: andrewcreagh.com/blog/2022/WeaR…

You couldn't make it to #NeurIPS2022 this year? Nothing to worry - I curated a summary for you below focussing on key papers, presentations and workshops in the buzzing space of ML in Biology and Healthcare 👇

Fab work with Emmanuel (Manos) Stamatakis Matthew Ahmadi Jason Gill Mark Hamer All made possible by the wonderful UK Biobank accelerometer collection - a transformational resource to identify if/how wearables are associated with major disease outcomes.

Is longer sleep duration associated with lower risk of Alzheimer's Disease? tinyurl.com/sleepPRS New preprint using wearables & genomics where polygenic score shows ~2% sleep variance & association with AD. Neat work led by Oxford Population Health (OxPop) early career fellow Angel Wong.

Pleased to share our preprint on an open-source method that offers the most robust way to measure steps from wrist-worn accelerometers in large-scale studies such as the UK Biobank. medrxiv.org/content/10.110… 🧵

Enjoy a drink while hearing about digital health and the wearable technology revolution! We are at the dawn of a digital healthcare revolution. The tsunami of physiological and behavioural data that can be collected from smartphones, smartwatches, fitness trackers and wearable

Hospitals are limited in data they can share with AI researchers for privacy reasons- Today, we're pleased to share one of the first uses of Federated Learning in the NHS! We trained & evaluated AI models across 4 UK hospitals without moving patient data medrxiv.org/content/10.110…

On a warm Monday evening, #pint23 is officially on! James St Tavern hosting Andrew Creagh and Becky Carlyle, about resilience in our brains and how wearables allow us to peek inside our own data and how to harness it Pint of Science UK

Great talk by Andrew Creagh for Pint of Science UK #pint23 tracking our brains and our gains 🧠💻📊

Job alert - research position for those interested in analysing and validating data from wearables Big Data Institute Oxford Population Health (OxPop) University of Oxford tinyurl.com/wearablesPost Please feel to email me to find out more about this position (closing date = 30th August).

New article! Read Digital health technologies and machine learning augment patient reported outcomes to remotely characterise rheumatoid arthritis dlvr.it/T2d1yf in npj Digital Med

Great work by Andrew Creagh in collaboration with OxWearables, GSK, Big Data Institute and Engineering Science, Oxford Behaviour recognitions models trained using research data (UK Biobank, and paperswithcode.com/dataset/captur…) work nicely on consumer devices (Apple Watch)!

Happy to announce our foundation model for wearables published at npj Digital Medicine today. This model sets a new standard in #Wearables, significantly outperforming human activity recognition benchmarks in diverse conditions. 🚀 nature.com/articles/s4174…

9🏅 Self-supervised learning for human activity recognition using 700,000 person-days of wearable data nature.com/articles/s4174…