0
0
As older adults increasingly prefer to age in their own homes and remain active within their communities, malnutrition often stands in the way of these plans. Conditions such as involuntary weight loss, micronutrient deficiencies, and frailty not only diminish quality of life but also frequently result in the need for long-term care facilities.
To address this growing issue, researchers at the University of Waterloo have developed an innovative AI-powered application designed to tackle malnutrition in aging populations. The app can accurately measure a person’s caloric and nutrient intake simply by “watching” what they eat—without requiring users to manually input data or upload photos.
Unlike traditional nutrition apps, which depend on users to log every meal and snack, this new app automates the process entirely. Once seated at a table with food, the user only needs a video camera mounted on a hardware device. The system detects when eating begins and tracks every bite that goes into the user’s mouth, making the process seamless.
Behind the scenes, a sophisticated visual language model (VLM) analyzes each frame of video, interpreting the type of utensil being used and the various stages of eating, including the opening of the mouth and chewing. Powered by Meta AI’s Segment Anything Model (SAM), the technology ensures accurate nutrient analysis by focusing precisely on what is being consumed and how much.
Dr. Yuhao Chen, a research assistant professor in the Department of Systems Design Engineering’s Vision and Image Processing (VIP) Lab, explains the advantage of this approach: “The more an app requires user input, the more likely it is that people won’t continue using it long term. We want to support ongoing nutrition monitoring, so we’re using AI to automate processes and place fewer requirements on the user.”
Initially, the system has proven highly effective for analyzing foods eaten with a spoon. The research is now expanding to include more complex eating utensils, such as forks, chopsticks, and even hands. The team is also looking to enhance the technology to monitor a broader range of substances impacting senior health, including supplements and medications.
Chen envisions a future where the app goes beyond just food analysis, saying, “The vision is to analyze whatever a person ingests, including supplements and drugs.”
Ease of use is a key component of the project. As Chris Czarnecki, a recent master’s graduate who contributed to the project, notes, “Some people aren’t comfortable using technology, so we want to make our application very easy to use.” Potential interfaces could include a smartphone app utilizing the built-in camera or even wearable glasses that seamlessly track food intake.
The ultimate goal is to empower aging adults to achieve optimal nutrition in the comfort of their own homes. By commercializing a solution that helps seniors live independently while staying nutritionally healthy, the team hopes to make a significant impact.
This AI-driven system is part of Waterloo’s broader research on healthy aging, addressing the growing concerns of malnutrition among older adults. With the number of people aged 65 and over expected to double in the next 30 years, the risk of widespread malnutrition could strain long-term care and healthcare systems. This groundbreaking technology represents a step toward helping seniors thrive as they age.
