Abstract:Background Ambulatory electrocardiogram (ECG) monitoring is essential for detecting paroxysmal cardiac events and tracking long-term physiological parameters. However, traditional Holter devices are cumbersome, restrict movement, and can cause skin irritation, resulting in sub-optimal patient compliance during extended monitoring periods. Wearable devices utilizing non-conventional form factors are emerging as promising non-intrusive alternatives for ECG monitoring. Purpose To determine the optimal location for single-lead ECG acquisition on the upper arm and assess the feasibility of using dry electrode armbands for cardiac monitoring when coupled with robust ECG processing software. Methods Stationary ECGs from 20 healthy adult participants were recorded across a two-phase data collection protocol and processed with clinically validated ECG software. In the first phase, the average signal amplitudes of wet-electrode ECGs obtained from 30 different configurations across the mid-brachial and proximal-brachial regions of the upper arm were assessed. In the second, the performance of two proprietary dry-electrode armbands, featuring either a single-point or three-point electrode strap, were evaluated on the optimal upper arm configuration using QRS detection and heart rate (HR) metrics. Comparative analysis was performed throughout using data simultaneously acquired on a gold standard Holter device in modified lead III (MLIII) configuration. Performance metrics were calculated relative to manually generated signal annotations. Results Wet-electrode-acquired ECGs from the proximal-brachial region exhibited significantly higher average signal amplitudes (0.12 ± 0.06 mV) compared to those from the mid-brachial region (0.05 ± 0.04 mV) of the upper arm (p < 0.05). The optimal configuration (A-F), positioned at the lateral and medial extents of the proximal-brachial region, demonstrated comparable average signal amplitudes (0.16 ± 0.07 mV, 0.16 ± 0.06 mV, 0.19 ± 0.08 mV) for single-point dry, three-point dry, and wet-electrode setups, respectively. Upon processing with ECG software, data from optimal dry-electrode configurations showed excellent average QRS sensitivity and specificity of 98.2% and 98.9% for the one-point strap, and 98.9% and 99.8% for the three-point strap, respectively. HR metrics from the optimized dry-electrode setups were comparable in accuracy to those derived from the reference Holter device, with the average HR error being similar across the three-point dry electrode strap (0.57% ± 0.28%), one-point dry electrode strap (1.16% ± 1.46%), and Holter device (0.74% ± 1.50%). Conclusion(s) Optimized upper arm single-lead ECG with dry electrodes can yield signals with sufficient amplitude and quality for automated ECG analysis. Coupled with robust ECG signal processing software, the cardiac metrics obtained from the optimized upper arm setups exhibited comparable performance to those derived from medical Holter data.

Heartbeat detection from high-density EMG electrodes on the upper arm at different EMG intensity levels using Zephlet

Automated Heartbeat Classification Exploiting Convolutional Neural Network with Channel-Wise Attention

A novel technique for the detection of myocardial dysfunction using ECG signals based on CEEMD, DWT, PSR and neural networks

Doppler Cardiogram: A Remote Detection of Human Heart Activities

Knowing Your Heart Condition Anytime

A High-Performance System for Weak ECG Real-Time Detection

A Heart Beat Rate Detection Framework Using Multiple Nanofiber Sensor Signals

Evaluating the feasibility of upper arm ECG for cardiac monitoring

Heart rate detection method based on Ballistocardiogram signal of wearable device：Algorithm development and validation

ECG R-wave Peaks Marking with Simultaneously Recorded Continuous Blood Pressure

Accurate Extraction Of Heartbeat Intervals With Near-Field Coherent Sensing

Beat-Level Interpretation of Intra-Patient Paradigm Based on Object Detection

A novel machine learning-enabled framework for instantaneous heart rate monitoring from motion-artifact-corrupted electrocardiogram signals

Robust heartbeat detection using multimodal recordings and ECG quality assessment with signal amplitudes dispersion

Radar-Beat: Contactless beat-by-beat heart rate monitoring for life scenes

An Energy-Efficient ECG Processor Based on HDWT and a Hybrid Classifier for Arrhythmia Detection

Efficient Extreme Learning Machine (ELM) Based Algorithm for Electrocardiogram (ECG) Heartbeat Classification

Artificial Intelligence-Enabled ECG Algorithm Based on Improved Residual Network for Wearable ECG

Electromechanical Dynamics of the Heart: A Study of Cardiac Hysteresis During Physical Stress Test

A novel technique for the detection of myocardial dysfunction using ECG signals based on hybrid signal processing and neural networks

Heart beat detection using a multimodal data coupling method