Over the last ten or so years, researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have been developing artificial intelligence (AI) x-ray vision. Earlier this month, the CSAIL researchers announced the RF-Pose, which is a form of technology using AI that teaches wireless devices how to track a person’s movement through a wall.
Being able to monitor how someone moves could offer important feedback to entities like medical providers on how patients with illnesses such as Parkinson’s and multiple sclerosis (MS) are coping with their conditions. This method could also benefit elderly people who want to continue independently living in their own homes, while still being monitored for injuries like falls, along with any critical or sudden changes in their activity.
“We’ve seen that monitoring patients’ walking speed and ability to do basic activities on their own gives healthcare providers a window into their lives that they didn’t have before, which could be meaningful for a whole range of diseases,” says Dina Katabi, who co-authored a paper that will be presented at the end of June during the Conference on Computer Vision and Pattern Recognition (CVPR) in Salt Lake City, Utah. “A key advantage of our approach is that patients do not have to wear sensors or remember to charge their devices.”
The MIT researchers are collaborating with doctors on various healthcare applications. The technology utilizes a neural network to analyze radio signals that deflect off someone’s body, to outline an individual’s figure and movements.
“Just like how cellphones and WiFi routers have become essential parts of today’s households, I believe that wireless technologies like these will help power the homes of the future,” says Katabi.
Outside of monitoring elderly individuals with medical conditions, RF-Pose can potentially be applied to other life-saving processes like search-and-rescue efforts to locate trapped survivors at disaster scenes. The research team also alleges they can use wireless signals for identifying a particular individual from a lineup of 100 people with up to 83 percent accuracy. To give the user more control over this technology, the research team plans to implement a consent mechanism, where the individual installing the device would engage a specific set of movements so RF-Pose can start monitoring its surroundings.
Filed Under: AI • machine learning, M2M (machine to machine)