Abstract:The most common non-invasive diagnostic model is the Electrocardiogram (ECG), which records the heart's electrical activity over time and is used to diagnose various heart conditions. Due to the requirements of a typical eHealth system, it is necessary to compress ECG signals for long-term data recording and remote transmission. Moreover, cardiovascular diseases (CVDs) have been considered the most long-lasting disorders in recent years. The transmission of information from the patient to the distant hospital is necessary because rapid analysis and treatment are essential for the condition to be cured. Also, the data must be in the form of lossless and high-predictability data. So, the goal of this study was to create a two-stage lossless Integer Adaptive Predictor (IAP) compressor that could be implemented on a Field Programmable Gate Array (FPGA) without introducing any data loss during the compression process. Before compression, the ECG signals are denoised using a Fast Normalized Least Mean Square (FNLMS) algorithm-based adaptive filter, which removes the undesirable noise presented in the signal. Here, the adaptive filter is designed based on the hybrid systolic folding structure and compressor-based multiplier architecture to minimize the power, delay and area consumption of the filter while performing the signal-denoising process. Xilinx and MATLAB are used to run simulations using the MIT-BIH Arrhythmia and PTB diagnostic databases. Several performance parameters are used to assess the proposed design's efficacy, and the results are compared to those of similar current designs. Consequently, the proposed compressor achieves a 45.23% compression ratio (CR) on MIT-BIH and a 10.87% average CR on the PTB diagnostic database, which demonstrates that the compression proficiency of the proposed design is high.

An Adaptive Real-Time ECG Data Compression Algorithm

Signal Adaptive Threshold for ECG Signal Compression Using False Discovery Rate Approach

ECG data compression based on wave atom transform

A Compression Method for ECG Based on DCT of Difference between Two Adjacent Frames

A Lossless Electrocardiogram Compression System Based on Dual-Mode Prediction and Error Modeling.

ECG Compression Based on Combining of EMD and Wavelet Transform

A Dual-Mode ECG Processor with Difference-Insensitive QRS Detection and Lossless Compression

[A New Compression Method for ECG Data].

ECG Data Compression Based on the Combination of Template Matching and Improved Quick Fitting of LADT

ECG Signal Compression and Optimization in Remote Monitoring Networks

Adaptive ECG Compression Scheme with Prior Knowledge Support Based on Compressive Sensing

A Highly Adaptive Lossless ECG Compression ASIC for Wireless Sensors Based on Hybrid Gomlomb Coding.

A Compression Algorithm for ECG Data Using Variable-Length Classified Template Sets

Real-time Lossless ECG Compression for Low-Power Wearable Medical Devices Based on Adaptive Region Prediction

An improved ECG data compression scheme based on ensemble empirical mode decomposition

FPGA-enabled lossless ECG signal compression system using an integer adaptive compressor

A Real Time QRS Detection Algorithm Based on ET and PD Controlled Threshold Strategy

A novel ECG compression algorithm using Pulse-Width Modulation integrated quantization for low-power real-time monitoring

A new automated compression technique for 2D electrocardiogram signals using discrete wavelet transform

Electrocardiograph Compression Based on Sifting Process of Empirical Mode Decomposition

A 2.63 μW ECG Processor With Adaptive Arrhythmia Detection and Data Compression for Implantable Cardiac Monitoring Device