Wearable devices could someday help people manage disabilities and debilitating diseases such as Parkinson’s Disease, but making them user friendly for patients has been a continuing challenge for researchers.
Ganapati Bhat, assistant professor in the School of Electrical Engineering and Computer Science, recently won the 2022 Outstanding Dissertation Award from the Association of Computing Machinery Special Interest Group on Design Automation (SIGDA) for his work to improve the devices. Bhat, who received his PhD at Arizona State University in 2020, will receive the award at the Design Automation Conference in San Francisco in July.
While millions of people live with physical disabilities, their diagnosis, treatment, and rehabilitation generally depend on what doctors see in the doctor’s office.
“After the patient leaves the clinic, there is no standard approach to continuously monitor the patient and report potential problems,” Bhat said. “The quality of life of this population could be improved significantly with the help of wearable internet-of-things (IoT) devices that combine sensing, processing, and wireless communication capabilities.”
In his dissertation, Bhat worked to solve several challenges that have prevented the devices from being more widely used. For people with Parkinson’s disease or who are recovering from a stroke, doing the physical work of plugging and unplugging batteries for charging can be cumbersome.
“We started thinking of how we can make the devices better in terms of the usability and have energy harvesting built into it, so that we don’t have to recharge them so often,” he said.
He improved energy use of the devices and developed an ultra-low-energy accelerator, so that the device could run on low energy provided by ambient light. The accelerator that he helped to develop is able to recognize how people are moving, read the data, and then classify the activity.
Bhat also improved activity recognition of such devices. So, for instance, everyone has different walking gaits, and these will change in people as an illness progresses. Bhat developed methods for a device to quickly learn about a new user, so that the device could better monitor a patient’s recovery or decline. The online learning framework that the researchers developed was able to improve the accuracy of activity recognition for new users by as much as 30%.
Bhat has published three papers based on the research and is continuing the work. He hopes to begin working in the next few years with patients and doctors to begin testing the technology under real-world circumstances.
“Widespread adoption of wearable devices for health monitoring could lead to a reduction of healthcare costs and better public health outcomes,” he said. “My main goal is trying to make an impact and make a difference in people’s lives.”