Automated food tracking system developed for nursing homes

February 4, 2022

Long-term care (LTC) or nursing homes face a common problem: malnutrition – it is estimated that more than half of its residents are either malnourished or at risk of malnutrition because of their refusal to eat certain foods. New technology which digitally tracks how much food residents consume was developed jointly by researchers at Canada’s University of Waterloo (UW), Schlegel-UW Research Institute for Aging, and University Health Network to reduce malnutrition and improve overall health in these facilities.

The artificial intelligence (AI)-based software compares photos of platefuls of food served, one taken before the meal and another once the resident has finished eating. By analysing factors such as the colour and depth of the food that’s left on the plate, the software can determine how much of each food type was consumed, and thus how much nutrition the resident received.

“Our system is linked to recipes at the long-term care home and, using artificial intelligence, keeps track of how much of each food [carbohydrates, proteins, etc.] was eaten to make sure residents are meeting their specific nutrient requirements,” said UW postgraduate engineer Kaylen Pfisterer.

Currently, the process of manually tracking food consumption at LTC homes by caregivers show an error rate of 50% or more. The food tracking software, by comparison, is accurate to within 5% inclusive of “fine-grained information on consumption patterns.”

“My vision would be to monitor and leverage any changes in food intake trends as yellow or red flags for the health status of residents more generally and for monitoring infection control,” said Pfisterer.

Pfisterer and colleagues at UW collaborated with personal support workers, dietitians, and other LTC care workers to develop the system, which would ideally be added to tablet computers used by front-line staff at the homes.

Read: Covid-19 vaccine trials excluded nursing mothers, says org

Health Care Asia Home