Research Grants

  1. Principal Applicant – Using in-home sensors to detect the physiological, functional, and behavourial changes related to ingesting cannabidiol oil in older adults with dementia living at home (Nominated Principal Applicant - Dr. Charlene Chu), CIHR SPOR, 2020-21, $90,300.
  2. Co-PI – Developing an at-home sensor system to detect social isolation and functional decline in high-risk older adults in the community (PI - Dr. Charlene Chu), CABHI SPARK, 2019-20, $49,961.
  3. Co-PI – Artificial Intelligence for the detection of anomalous events in long term care (PI - Andrea Iaboni), CABHI SPARK, 2019-2020, 49,972.
  4. PI – Location tracking-based health status indices and their integration into clinical support tools in long-term care., (PI - Andrea Iaboni), AGEWELL NCE Catalyst CRP Funding, 2019-2020, $28,500.
  5. Co-Applicant – Intelligent intervention system to deter older people (living with mild dementia) from physical inactivity, (PI - Kristine Newman), Ryerson FCS Seed Grant, 2018-19, $6,000
  1. CareBand: Wearable technology for people with dementia, (PI - Adam Sobol), NIH SBIR Grant, 2019-2020, US $228,778

Research Projects

Providing care for a rapidly aging population, especially people living with dementia (PLwD), constitutes a major challenge for global healthcare. Older adults with advanced dementia living in long-term settings often have behavioural and psychological symptoms of dementia, with agitation and aggression amongst the most common symptoms. These behaviours can lead to resident-on-resident violence and workplace violence towards staff in long-term care facilities. In this project, our goal is to develop an automated system that can alert staff and provide personalized interventions when episodes of agitation occur by detecting, tracking over time, and predicting such behaviours among PLwD. We are using a multi-modal sensor network that collects data using physiological and motor sensors, video cameras, motion/door sensors and pressure mats to build novel machine/deep learning classifiers.

A fall is an abnormal activity that occurs rarely, infrequently and diversely. Therefore, it is difficult to collect training data for falls. Traditional machine learning classifiers may not work well in such skewed scenario. We are taking an alternative approach to only train classifiers on normal activities that are present in abundance and flag a fall as an abnormal activity. We are using vision based non-invasive sensors that protect privacy, such as thermal camera, depth camera and near-IR camera to test our models. We are developing models using 3D Spatio-Temporal Convolutional Autoencoders (3D-STCAE) to achieve this task. A short demo to detect an unseen fall using 3D-STCAE is shown below.

An Intelligent Assessment system is being developed to detect Social Isolation, Functional and Cognitive decline among older adults. This system is a zero-effort assistive technology solution that utilizes low cost smart devices in a home setting and requires little or no effort from the people using it. The system will detect prolonged periods of inactivity, motion and other physiological and ambient indicators suitable for the task. The goal of this system is to develop a passive assistive system that does not aggressively prompts the person using it, rather adaptively learns their daily behaviour. A survey on the role of technology for detecting social isolation is currently being developed.


Team Members

  • Nizwa Javed, Research Analyst, 2020
  • Katherine Rich, Undergraduate Student, IBBME, U of Toronto, 2020
  • Sepehr Rashidi, Volunteer Researcher, 2020
  • Logan Rooks, Undergraduate Thesis Student, Engineering Science, 2019-
  • Matthew Nogas, Technical Analyst (2018-2019), Undergraduate Trainee (2018)
  • Thaejaesh Sooriyakumaran, Undergraduate Summer Intern, 2019
  • Brandon Malamis, Undergraduate Summer Intern, 2018, 2019
  • Paris Rosarie, Undergraduate Summer Intern, 2019
  • Jacob Nogas, Undergraduate student (PEY year), 2017-2018
  • Tong (Maggie) Zhu, Undergraduate Summer Intern, 2017

Research Interests