Damiano Zanotto (Stevens Institute of Technology)
18 Mar. 2020. An engineering lab developed an electronic shoe insole that gathers data for analysis with artificial intelligence to accurately measure a person’s gait patterns. The lab’s senior author, Damiano Zanotto, mechanical engineering professor at Stevens Institute of Technology, is also a recipient of a Faculty Early Career Development grant from National Science Foundation to advance his research.
Zanotto’s Wearable Robotic Systems Lab at Stevens Tech in Hoboken, New Jersey, studies intelligent systems that monitor or assist in human movement. One the lab’s projects is SportSole, a shoe insole with electronic sensors to accurately capture data in real time on a person’s gait, or patterns of walking or running. While technology for capturing these data are advancing, accuracy of the data captured, say Zanotto and other authors of a recent paper, are either not reliable or too expensive to collect in everyday use.
The researchers describe the SportSole system in the January 2020 issue of the journal IEEE Transactions on Neural Systems and Rehabilitation Engineering (paid subscription required). That system places inexpensive and off-the-shelf sensors into the SportSole to capture data on movement and pressure. The device then provides data on a host of factors affecting a person’s gait patterns, such as length of stride, ground clearance, foot trajectory, cadence, and walking speed, at a rate of 500 readings per second.
The Stevens Tech team then wrote machine learning algorithms to analyze the data and provide summary analytics of a person’s walking and running patterns. The compact algorithms are based on support vector regression models, statistical formulas for fitting captured data within allowable boundaries, that fit on a built-in micro-controller, allowing for real-time analysis. Tests with 14 volunteers running and walking for six minutes show the SportSole system calculations highly correlate with reference stationary lab equipment for stride length, velocity, and foot clearance.
“We’re now able to accurately analyze a person’s gait in real time, in real-world environments,” says Zanotto in a Stevens Tech statement. He adds, “We’re achieving the same or better results at a far lower cost, and that’s a big deal when it comes to scaling this technology.” The university says the researchers applied for two patents on the technology, which is attracting commercial interest from several companies and professional sports franchises.
Earlier this month, Zanotto received a grant from National Science Foundation to advance his lab’s work on a robotics-assisted ankle brace for rehabilitation. The Faculty Early Career Development or CAREER award of nearly $600,000 supports development of an orthotics system on the ankle that provides real-time reinforcement learning to help people in rehab from a stroke or other neurological conditions regain their walking ability. The device is also expected to help people with gait impairments navigate unknown terrains with unforeseen obstacles.
More from Science & Enterprise:
- Infant Jumpsuit Captures Objective Movement Data
- Wireless Smart Bandage Designed for Chronic Wounds
- Skin Patch Senses Glucose, Emits Insulin as Needed
- FDA Clears A.I. Algorithms for Stethoscopes
- Ingested X-Ray Capsule Shown Safe, Feasible
* * *
RSS - Posts
You must be logged in to post a comment.