Amazon and Seattle-based University of Washington researchers came up with an artificial intelligence-powered system that would use sonar technology to pick up vibrations caused by chest wall movements. The speakers would emit audio signals into a room at a volume that humans cannot hear. An algorithm would then use bouncing pulses to identify patterns from a human’s chest wall, and a second algorithm would determine the amount of time between two heartbeats.
The researchers said physicians could use the smart speakers in telemedicine appointments to get data that would otherwise require wearable equipment.
Researchers tested the software on 26 healthy participants and 24 hospitalized participants with varying cardiac conditions. They compared smart speaker findings to the results from ECG monitors and found that the results were relatively accurate. Differences in readings varied by an amount that wasn’t medically relevant, the Post reports.
More articles on health IT:
Google’s 9 most recent health-related job openings
Ascension’s innovation arm launches $285M venture fund
3 recent health IT exec moves