ATS 2024: Validation of an Acoustic AI Cough Monitoring System
Summary
Over the 477 hours monitored, 4102 cough-seconds were recorded from the 20 subjects; the CoughMonitor detected 3715 of these (overall sensitivity 90.57%) and had 520 false positives (overall false positive rate of 1.09 false positives per hour).
The ability to passively and continuously monitor cough would significantly improve cough management andresearch. Recent advances in acoustic AI have prompted the development of automated cough monitoring technology. Here wedescribe for the first time the process and results of validating an acoustic AI based cough monitor
