Electrocardiographic Signal Compression Based on Distributed Source Coding

Electrocardiogram (ECG) signals can be used to monitor and diagnose heart-related diseases. Usually, they are obtained with specific mobile equipment or in a hospital, depending on the available storage and transmission infrastructures. Moreover, they may be acquired with many electrodes to identify specific conditions. However, since memory and bandwidth are finite resources, compression techniques become necessary when ECG data must be acquired, stored, processed, and transmitted. Additionally, it is paramount that mobile devices be simple and affordable, which is once again related to memory and transmission capabilities and ultimately affects the associated hardware infrastructure. This work tackles the mentioned problems and proposes a new ECG compression methodology based on distributed source coding concepts, whose main goal is to provide low-complexity encoders. Experiments with an actual implementation showed feasibility and effectiveness, even when comparing it with traditional approaches, as long as its aspects are aligned with an application’s specific conditions.