Single-lead ECG Compression for Connected Healthcare Applications

Abstract

Preventive healthcare is achievable through physiological long-term remote monitoring. In connected healthcare, wearables that can collect physiological signals such as electrocardiograms (ECG) and electroencephalogram (EEG) can help improve health outcomes in society. For single-lead ECG devices, there are still limitations for this role that includes short time continuous operability and uncomfortable sensors worn by the user making it non-appealing for uninterrupted remote monitoring. However, with the current advances in microelectronics, embedded systems, sensors, and Internet of Medical Things (IoMT), long-term monitoring is realizable. A decrease to the overall power consumption of the wearable leads to an increase in device longevity while dry ECG electrodes can be used to increase user comfort. This work proposes a lossless LempelZiv Welch (LZW) compression algorithm used to compress and optimize the raw ECG data obtained from a 3D printed dry electrode based single-lead ECG device. This approach utilizes the ECG's inherent waveform characteristics. The single-lead ECG's R-peak and RR-intervals are used as one-bit information that are further compressed for shorter wireless transmission, leading to an increase in battery life and device operation. The algorithm showed a high compression ratio (CR) for 10 seconds, 30 seconds, 1-minute and 5-minute ECG signals where CR was 0.99, 0.91, 0.91, 0.92, respectively. For the 5-minute ECG signal, the size of data decreased from 225 Kbytes to 18.75 Kbytes while retaining R-peak and RR interval information for heart rate (HR) and heart rate variability (HRV) calculations. This work adds to the current progress in single-lead ECG in long-term continuous remote monitoring for connected healthcare applications.