Date: Tuesday 23 September 2025

A webinar recording of the GOLTC Data Science Interest Group from 23 September, 2025

Agitation and aggression in people living with dementia are among the most distressing and difficult-to-manage symptoms in clinical settings. Traditional assessments rely on carer reports, often after the fact, limiting the opportunity for early intervention.

In this webinar, Dr Abeer Badawi presents her team’s groundbreaking work on a real-time system that combines physiological data from wearable devices with video-based AI to detect and even anticipate agitation episodes. Developed over five years and tested in a clinical pilot, in a partnership with Ontaria Tech University and Ontario Shores Hospital, the system identified pre-agitation patterns up to six minutes before behavioural symptoms emerged. The research opens up new possibilities for using digital monitoring to reduce distress.

The webinar will cover:

1. Introduction and welcome (Adelina Comas-Herrera and Sam Rickman, Care Policy and Evaluation Centre (CPEC), LSE)
2. Predicting and preventing agitation in dementia care using wearable sensors and video analysis (Dr Abeer Badawi, York University and Vector Institute, Toronto)
3. Discussion and questions (Chair: Tommy Henderson-Reay, NHS England)

 

This webinar is to discuss the following paper:

Badawi, A., Elmoghazy, S., Choudhury, S., Elgazzar, S., Elgazzar, K., & Burhan, A. (2024). A Novel Multimodal System to Predict Agitation in People with Dementia Within Clinical Settings: A Proof of Concept. arXiv preprint arXiv:2411.08882.

It will be presented by the first author, Abeer Badawi.