Heart rate monitoring using wrist-type photoplethysmographic signals during subjects' intensive exercise is a difficult problem, since the signals are contaminated by extremely strong motion artifacts caused by subjects' hand movements. So far few works have studied this problem. In this study, a general framework, termed TROIKA, is proposed, which consists of signal decomposiTion for denoising, sparse signal RecOnstructIon for high-resolution spectrum estimation, and spectral peaK trAcking with verification. The TROIKA framework has high estimation accuracy and is robust to strong motion artifacts. Many variants can be straightforwardly derived from this framework. Experimental results on datasets recorded from 12 subjects during fast running at the peak speed of 15 km/h showed that the average absolute error of heart rate estimation was 2.34 beat per minute, and the Pearson correlation between the estimates and the ground truth of heart rate was 0.992. This framework is of great values to wearable devices such as smartwatches which use PPG signals to monitor heart rate for fitness.
1 August 2015 in Biomedical Signal Processing and Control
In electrical impedance tomography (EIT), it is difficult to obtain the intracranial impedance due to the highly resistive skull enclosing the brain. Therefore, a high-precision data acquisition system is required for brain EIT. In this paper, we used a high-precision digital synthesis method and a digital demodulation technique with high noise immunity to eliminate random errors. Moreover, we focused on two problems encountered during EIT data acquisition: 1) the shunt effect on the excitation current due to the distributed capacitance between electrodes and ground and 2) high common-mode voltages in the boundary measurements. We designed a new electrode interface to reduce the influence of the distributed capacitance and a programmable current source to accurately compensate for the excited current. We also proposed a new voltmeter circuit with improved CMRR. Overall, this EIT data acquisition system can produce a programmable current with SNR greater than 89 dB. It can also measure the voltage difference precisely with CMRR higher than 75 dB with a 1-kQ impedance imbalance. The results on a calibration model show that this system has a high SNR of 83 dB and a low reciprocity error of 0.125%. In addition, EIT imaging results were acquired using a brain physical phantom. The system can detect small disturbances of 0.35% in volume (1.99% of the cross-sectional area) and 17% in resistivity. Experiments on healthy volunteers also suggest that small intracranial impedance variations due to temporary occlusion and reperfusion of the unilateral carotid artery may be monitored by the system.