Development of a Low-Power Heart Rate Monitor Device for Observation of Heart Rate Variability

摘要

Heart rate variability is a useful tool for observing the autonomous modulation from the nervous system on the heart rate. Current acquisition concept relies on the electrocardiogram with post-processing to retrieve heart rate signal and is inefficient for the intended purpose. The IPFM model presents a simple method for heart beat generation and can be modified to give a very efficient heartbeat sampler. In this paper, it is shown how a new acquisition concept can be obtained from the manipulation of the time varying threshold IPFM model. A low-power acquisition solution is proposed and implemented on an embedded platform as a proof-of-concept. Validation of the proposed system is shown by comparison with the electrocardiogram as a gold standard. Results show good validation performance on 3 test recordings of 5 min each ($gt1000$ beats total, aggregate MSE $=1.73imes 10^{-4}$). The sampled signal can also be used as a substitute QRS detector (93.16% detection rate on $gt1000$ beats). Another line of study aims at analyzing sampling rate reduction effect on observed heart rate variability, to assess potential reduction of power consumption. It was shown that the mean HRV profile can be reliably obtained even at low count frequencies, whereas higher frequencies are required for variable profile capture. The best frequency band is more than 2 times less than standard ECG recording rate.