Characterization of a novel heart and respiratory rate sensor.

Existing methods of electronic patient monitoring require tethering the patient to the device, which is not well tolerated. This study characterizes the performance of a novel sensor array and digital signal processing (DSP) algorithms that extract heart and respiratory rates. The sensor lies under the sheets on a hospital bed, and when the patient lies down, it detects pressure waves generated by the heart, and by the act of breathing. The algorithms identify these signals of interest, filtering out extraneous signals. Output of the algorithms was compared to output from ECG and transthoracic impedance, taken from the same subject, at the same time. Forty-four adult volunteers were recruited. The results demonstrated an average of mean differences for heart rate of 0.50 beats per minute, with a standard deviation of 0.51. The average of mean differences for respiratory rate was 0.39 with standard deviation of 0.55. These results suggest this noninvasive, non-restrictive method of measuring heart and respiratory rates may be a viable solution to the problem of decreased vigilance of patient condition faced on the in-patient wards. Future studies will characterize performance in ill populations, and examine alarm schemes that are both highly sensitive and specific for the target population.