New know-how may assist cut back malnutrition and enhance total well being in long-term care properties by routinely recording and monitoring how a lot meals residents devour.
The good system, developed by researchers on the College of Waterloo, the Schlegel-UW Analysis Institute for Getting old and the College Well being Community, makes use of synthetic intelligence software program to research photographs of plates of meals after residents have eaten.
The delicate software program, which examines color, depth, and different picture options, can estimate how a lot of every form of meals has been consumed and calculate its dietary worth.
“Proper now, there isn’t a solution to inform whether or not a resident ate solely their protein or solely their carbohydrates,” mentioned Kaylen Pfisterer, who co-led the analysis whereas incomes a PhD in techniques design engineering at Waterloo.
“Our system is linked to recipes on the long-term care residence and, utilizing synthetic intelligence, retains observe of how a lot of every meals was eaten to verify residents are assembly their particular nutrient necessities.”
It’s estimated that greater than half of residents of long-term care properties are both malnourished or liable to malnutrition.
Meals consumption is now primarily monitored by employees who manually report estimates of consumption by plates as soon as residents have completed consuming.
Robert Amelard, a Waterloo alumnus and postdoctoral fellow at College Well being Community, mentioned the subjectivity of that course of ends in an error fee of fifty per cent or extra. By comparability, the automated system is correct to inside 5 per cent, “offering fine-grained data on consumption patterns.”
Researchers collaborated with private assist employees, dietitians and different long-term care employees to develop the system, which saves time in addition to improves accuracy and would ideally be added to pill computer systems already utilized by front-line employees to maintain digital information.
“My imaginative and prescient could be to observe and leverage any modifications in meals consumption traits as yellow or crimson flags for the well being standing of residents extra typically and for monitoring an infection management,” mentioned Pfisterer, now a scientific affiliate on the College Well being Community Centre for World eHealth Innovation.
The analysis group additionally included Heather Keller, a professor of kinesiology and well being sciences, Alexander Wong, a techniques design engineering professor, and college students Audrey Chung, Braeden Syrnyk and Alexander MacLean.
A paper on their work, Automated food intake tracking requires depth-refined semantic segmentation to rectify visual-volume discordance in long-term care homes, seems within the journal Scientific Experiences.
Journal
Scientific Experiences
Disclaimer: AAAS and EurekAlert! aren’t liable for the accuracy of stories releases posted to EurekAlert! by contributing establishments or for the usage of any data by means of the EurekAlert system.
The post Food-tracking AI system developed to reduce m appeared first on Healthy DieTips.
from Healthy DieTips https://ift.tt/9A7RyIvou
via IFTTT
Comments
Post a Comment