Wearable Technology: Signal Recovery of Electrocardiogram from Short Spaced Leads in the Far-field Using Discrete Wavelet Transform Based Techniques

Abstract

Bipolar ECG leads recorded from closely spaced electrodes are challenging in any context. When they are positioned distally with respect to the source field (far-field), the recovery of clinically useful signal content represents an even greater challenge. Due to the increased interest in ambulatory wellness devices, particularly wrist-worn devices, there is a renewed interest in recovering ECG signals from distally located bipolar leads. In this study 10 bipolar leads were simultaneously recorded at various locations along the left arm. At the same time, a conventional proximal reading on the chest using Lead I was also recorded and stored. This process was repeated for 11 healthy subjects. ECGs were recorded for a period of approximately 6 minutes for each subject and sampled at a frequency of 2048 Hz. Wavelet-based filtering using Daubechies 4 wavelet decomposition and soft threshold was applied to each lead. QRS detection performance was assessed against Lead I for each subject. This investigation found that a lead positioned transversally (using BIS gelled electrodes) on the upper arm provided the best accuracy against the benchmark QRS detection (SEN = 0.998, PPV = 0.984). The most distally positioned bipolar lead using dry electrodes faired least favourable (SEN = 0.272, PPV = 0.202).

Armband Sensors Location Assessment for Left Arm-ECG Bipolar Leads Waveform Components Discovery Tendencies around the MUAC Line

“…The results from this study may be contrasted with the findings in a previous study carried out on a pilot patient database (N = 11) [ 27 ]. There, prior to QRS detection performance analysis with the PT algorithm, arm-ECG signal denoising was successfully achieved by an advanced Discrete Wavelet Transforms (DWT) ECG filtering technique using a 4th-order wavelet decomposition, with the db4 Daubechies wavelet as the mother wavelet.…”

“…There, prior to QRS detection performance analysis with the PT algorithm, arm-ECG signal denoising was successfully achieved by an advanced Discrete Wavelet Transforms (DWT) ECG filtering technique using a 4th-order wavelet decomposition, with the db4 Daubechies wavelet as the mother wavelet. Then, the performance analysis results [ 27 ] indicated that the bipolar lead on the upper-arm, defined as Ch8 − Ch10, presented the best sensitivity mean value (99.8%) and mean PPV (98.4%) performance. In contrast, the performance results obtained here indicated a maximum sensitivity mean value of 86.8% (with arm Lead-1).…”

“…Thus, a robust Single-Fiducial-Point (SFP) alignment technique was adopted for the SAECG process in this work; as adopted in previously reported works [ 7 , 12 , 15 , 17 ]; the SFP technique provides a single and highly deterministic time point with respect the ventricular depolarisation event in the ECG, derived from processing the stable Chest Lead I signal as the reference QRS event timing, thus permitting an accurate arm-ECG beat alignment process. Further analysis of arm leads QRS detection adopted the conventional Pan-Tompkins (PT) QRS detection algorithm [ 19 , 25 , 26 , 27 ]. It offers better results for evaluating bipolar arm lead sensitivity, precision (positive predictivity) QRS detection and HRV metrics performance attributes; when referred to true positive QRS events detected on the Chest Lead I ECG, which can be taken as the absolute true event annotation [ 27 ].…”

“…Further analysis of arm leads QRS detection adopted the conventional Pan-Tompkins (PT) QRS detection algorithm [ 19 , 25 , 26 , 27 ]. It offers better results for evaluating bipolar arm lead sensitivity, precision (positive predictivity) QRS detection and HRV metrics performance attributes; when referred to true positive QRS events detected on the Chest Lead I ECG, which can be taken as the absolute true event annotation [ 27 ]. Both of these approaches (SFP and PT) were implemented using embedded signal adaptive processes for moderately fast (within 6 heart beats) appropriate adjustment (adaptation) of amplitude and time thresholding ECG parameters [ 26 ].…”

“…The Pan-Tompkins (PT) algorithm uses the slope, amplitude, and width of an integrated window for the identification of QRS complex events in the ECG [ 19 , 25 ]. Amplitude and pulse width thresholding parameters in the PT algorithm decision stage were set specifically for discriminating and validating QRS complexes of sinus beats; rejecting potential noise peaks in the algorithm output, reducing the false events detection rate produced by non-ECG signal components and artefacts present in the Chest Lead I and arm-ECG signals [ 26 , 27 ]. To adapt to slow variations in QRS amplitude and duration, respective threshold variables were designed to adjust automatically at every positive decision stage (valid QRS event detection).…”