Alzheimer's Assocation Research only
All of
  • Go to
  • Research Center
  • AAIC
  • Journal
  • Grants
  • TrialMatch
  • Press
  • Donate
  • Contact Us
Science and Progress
Clinical Trials
Funding and Collaboration
You can Help
Stay Current
Video and Resources

Text Size

Small text Medium text Large text

Research Grants 2015

To view an abstract, select an author from the vertical list on the left.

2015 Grants - Taati

The Automated Monitoring of Gait as a Predictor of Fall Risk

Babak Taati, Ph.D.
Toronto Rehabilitation Institute-UHN
Toronto, Ontario, Canada

2015 New Investigator Research Grant

Can a novel technology that detects subtle changes in walking patterns be used to predict the risk of fall in people with Alzheimer’s?

People with dementia are at a high risk for falls which are a leading cause of injury and can contribute to loss of independence and quality of life. Many falls could be prevented if there was a way to predict an individual’s likelihood of falling and offer interventions to reduce this risk. New research suggests that subtle changes in a person’s gait, or manner of walking, may be linked to the risk of fall in people with dementia. A tool that can detect these changes could be used to monitor people with dementia and reduce their risk of injury.

Research Plan
Babak Taati, Ph.D., and colleagues developed a tool (ambient mobility, balance, and gait evaluation and monitoring technology; AMBIENT) that measures walking patterns using a wall-mounted camera. AMBIENT can monitor the mobility of people with dementia in a way that is accurate, non-intrusive and cost-effective. The researchers can use this technology to assess walking patterns and stability of older adults with dementia as they move about their environment.

For their current work, the researchers will conduct a small pilot study using AMBIENT to monitor fall risk in older adults in a hospital dementia care unit. AMBIENT will collect daily information about individuals’ walking patterns and their stability. The researchers will also analyze any falls experienced while hospitalized to identify patterns of mobility and changes in these patterns that occur prior to a fall. They will use advanced data analysis methods called “machine learning” to combine these patterns into a “predictive model.” They will determine if the model can detect specific changes in mobility occurring in the 7 days before a fall.

The results of this study will provide the information needed for the researchers to conduct larger human clinical trials using AMBIENT to detect and predict fall risk. This novel technology has the potential to prevent injury and prolong independence and quality of life for individuals with dementia.

Co-funded by the Alzheimer’s Association and Brain Canada Foundation

Alzheimer's Association International Conference | July 16-20, 2017, London, England

Abstract Submissions Now Open

The Scientific Program Committee is now accepting submissions for poster
presentations, oral presentations and featured research sessions.