High-order contrasts for independent component analysis
Neural Computation
Principal component analysis in ECG signal processing
EURASIP Journal on Applied Signal Processing
Application of independent component analysis in removing artefacts from the electrocardiogram
Neural Computing and Applications
The smart car seat: personalized monitoring of vital signs in automotive applications
Personal and Ubiquitous Computing
Fast and robust fixed-point algorithms for independent component analysis
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Non-contact capacitive ECG measurements (cECG) have applications in various unobtrusive and ubiquitous systems. However, cECG signals are frequently corrupted by interference and motion artifacts. In this work array processing methods, such as blind source separation, were used to reduce the impact of motion artifacts on QRS detection. The capacitive sensor array was integrated in a bed mattress and covered with two insulating sheets. The array processing methods were compared in terms of their QRS detection error rates (De). Results of our study with five healthy subjects in different recording conditions showed that, when using array processing methods, QRS detection performance during body motion can be substantially improved (De reduced from 0.46 on raw sensor data to 0.06 for a channel difference method). We concluded that array processing is a promising approach to achieve motion-resistant QRS detection and thus suggest wider use of capacitive sensor arrays.