LoCLoP: Low-cost/Low-processing Power Methodology for Deriving Heart Rate and Respiratory Rate in Time-critical Domain

Abstract

This paper proposes a methodology (LoCLoP) for deriving the heart rate and respiratory rate vital signs from a given ECG signal, leveraging affordable low-cost equipment which can be utilized by urgent medical units or even volunteers in events of considerable number of injured civilians. The LoCLoP methodology has been evaluated using more than 400 ECG signals from two distinct ECG data sets. Its robustness is measured in terms of accuracy estimation (sensitivity of 0.94/0.74, and positive predictivity of 0.92/0.93), accuracy in the triage process (sensitivity of 0.74, and positive predictivity of 0.95), and time performance (the average-case execution time is 0.02 seconds). The implemented LoCLoP methodology suggests an optimal ECG signal length (30 seconds) for producing sufficiently accurate heart rate and respiratory rate estimations.