No abstract

The capacitive electrocardiograph (cECG) has been tested for several measurement scenarios, including hospital beds, car seats and chairs since it was first proposed. The inferior signal quality of the cECG compared to the gold standard ECG guides the ongoing research in the direction of out-of-hospital applications, where unobtrusiveness is sought and high-level diagnostic signal quality is not essential. This study aims to expand the application range of cECG not in terms of the measurement scenario but in the profile of the subjects by including subjects with implanted cardiac pacemakers. Within this study, 20 patients with cardiac pacemakers were recruited during their clinical device follow-up and cECG measurements were conducted using a seat equipped with integrated cECG electrodes. The multichannel cECG recordings of active unipolar and bipolar pacemaker stimulation were analyzed offline and evaluated in terms of Fβ scores using a pacemaker spike detection algorithm. Fβ scores from 3652 pacing events, varying from 0.62 to 0.78, are presented with influencing parameters in the algorithm and the comparison of cECG channels. By tuning the parameters of the algorithm, different ranges of Fβ scores were found as 0.32 to 0.49 and 0.78 to 0.88 for bipolar and unipolar stimulations, respectively. For the first time, this study shows the feasibility of a cECG system allowing health monitoring in daily use on subjects wearing cardiac pacemakers.

No abstract

Unobtrusive measurement technologies for vital signs, such as capacitively coupled electrocardiography (cECG), allow for health monitoring outside the clinical domain, for example in automotive applications. While cECG has the potential to deliver accurate information about the driver's heart, the signal quality is very volatile compared to standard ECG and can be corrupted easily by motion artifacts. In this work, we present a signal-level fusion algorithm based on a convolutional neural network (CNN) to locate individual heartbeats in three-channel cECG signals. To design and optimize the network's structure, data from the PhysioNet / CinC challenge "Robust Detection of Heart Beats in Multimodal Data" was used as an independent source. To train and test the algorithm, we used cECG data from six subjects and three different driving scenarios (highway, city, and proving ground) that is freely available as part of our UnoViS-database. Data consisted of 31 recordings with a total duration of 13.4 hours. Leaveone-subject-out cross validation was performed to assess the algorithm's performance. We achieved a sensitivity of 88.0% and a positive predictive value of 95.2% compared to the reference ECG, with a root-mean-square R-Peak localization error of 20.81 ms. The developed algorithm is available for download via the UnoViS-Website.

No abstract