The system, called Sapple, collects data from two separate sensors: a wall-mounted device that tracks the movements of residents, and a smart electricity meter that monitors the total energy consumed in the home. This data is fed to a deep learning model that connects the resident’s location to the energy signals to work out how they’re using microwaves, stoves, and hair dryers. [Read: Body-worn sensors to predict COVID-19 infections in US military hospitals] Sapple was created by MIT Computer Science and Artificial Intelligence Lab (CSAIL) researchers, who. They say that combining the two sensors allows the system to detect when an appliance is used without needing any human-labeled data.
Healthy habits
MIT PhD student Chen-Yu Hsu, the lead author of a paper on Sapple, said the tech could be used anywhere with a utility meter: The researchers say the system can spot if elderly people are struggling to maintain healthy lifestyles. They envision healthcare professionals using the data to provide advice on personal hygiene, dressing, eating, maintaining continence, and mobility. But if you like to binge on ready meals without being told off, I’d recommend giving Sapple a miss.
