Derivation of Respiratory Signals from Single-Lead ECG

Abstract

Respiratory patterns carry clinically useful information. In many cases, ECG signals but not respirations are routinely monitored. We describe a signal processing technique which derives respiratory waveforms from ordinary single-lead ECG. First, we transform the original single-lead ECG signal using quadric B-spline wavelet, then extend R waves of supra ventricular beats in the 2nd approximation of wavelet transform. The extended signal is passed through a low-pass filter to reduce sample rate to 5 Hz, then through a band-pass filter. The output of the band-pass filter is the ECG derived respiration (EDR). We compare examples of EDR signals with conventional respiration measurements. The results show the EDR signals bear remarkable resemblance to the measured respiration. In many cases apneas are easily identifiable. This technique is applicable to real-time remote health care monitoring systems, requires no supplementary transducers or hardware.