Efficient electrocardiogram signal coding for medical devices using linear prediction

Q: What is the clinical evidence from this study?

Population: Electrocardiogram (ECG) signal compression for cardiac monitoring. Intervention: Linear predictive coding (LPC) with threshold-based residual sparsification (alpha=0.005) vs. LPC with exact (unthresholded) residual signal (alpha=0). Primary outcome: Compression ratio (CR) and percent root-mean-square difference (PRD).

Q: What does this research mean for the field?

Linear predictive coding combined with threshold-based residual sparsification achieves ECG compression ratios up to 38:1, and approximately 20:1 with a 5.03% PRD while preserving clinically relevant waveform features. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Sameer Ahmad Hasan; Samer Alabed; Amer Alsaraira; Yasin Omar

doi:10.11591/eei.v15i5.11688

October 1, 2026Bulletin of Electrical Engineering and InformaticsOpen Access

Efficient electrocardiogram signal coding for medical devices using linear prediction

View Full Paper

Ask AI

Bookmark

Share

Key result

Linear predictive coding with residual sparsification compresses ECG data up to ~38-fold while preserving waveform fidelity.

Why the study?

The growing adoption of wearable health monitoring and internet of things-based medical devices has increased the demand for efficient ECG compression techniques that reduce data size while preserving diagnostic fidelity.

Does a tenth-order linear predictive coding compression method with threshold-based residual sparsification improve compression ratios while preserving diagnostic waveform fidelity in ECG signals from PhysioNet databases?

Key Points

To develop a low-complexity electrocardiogram (ECG) signal compression framework using linear predictive coding (LPC) and residual sparsification that preserves diagnostic fidelity for wearable and IoT devices.
Modeled ECG signals from PhysioNet databases using a tenth-order LPC framework segmented into frames at a sampling rate of 1000 Hz.
Applied threshold-based residual sparsification across multiple alpha thresholds (0, 0.002, 0.005, and 0.007) to selectively suppress residual samples below threshold values.
Organized compressed LPC parameters and sparse residuals into fixed-size packets to evaluate reconstruction fidelity via signal-to-noise ratio (SNR) and percent root-mean-square difference (PRD).
Applying a moderate threshold of alpha = 0.005 yielded a compression ratio of 20.07:1 (around 20:1) with a PRD of 5.03% (PRD of 5%), global SNR of 13.33 dB, and average SNR of 15.19 dB while preserving the P wave, QRS complex, and T wave.
Increasing the threshold to alpha = 0.007 achieved a maximum compression ratio of 38:1, with a PRD of 8%, global SNR of 6.2 dB, and average SNR of 7.99 dB.
The scheme lowered the continuous transmission data rate to below one kbps, compared to conventional compressed rates of approximately 2–8 kbps.

Structured PICO

Does a tenth-order linear predictive coding compression method with threshold-based residual sparsification improve compression ratios while preserving diagnostic waveform fidelity in ECG signals from PhysioNet databases?

P

Population

ECG recordings from publicly available PhysioNet databases, segmented into frames at a sampling rate of 1000 Hz (evaluated on 3-second and 200-second ECG signal segments)

I

Intervention

Linear predictive coding (LPC) compression method (tenth-order LPC framework with line spectral frequency representation) combined with a residual thresholding mechanism (sparsified residual with threshold alpha = 0.002, 0.005, and 0.007) and 4-bit log2-based quantization

C

Comparator

Reconstruction using full (exact) unmodified residual signal (alpha = 0)

O

Outcome

Compression ratio (CR) and signal reconstruction fidelity measured by percent root-mean-square difference (PRD) and signal-to-noise ratio (SNR_global and SNR_avg)surrogate

Main Result

Absolute Event Rate: 20.07% vs 4%

A tenth-order linear predictive coding algorithm with residual thresholding achieves a 20:1 ECG signal compression ratio (reducing transmission rates to below 1 kbps) with a 5.03% reconstruction error, offering a low-complexity solution for wearable and IoT-based cardiac monitoring.

Limitations

Evaluation was conducted on a limited dataset.
Tested primarily on ECG signals with synthetic small white Gaussian noise rather than real-world artifacts like motion artifacts and baseline wander.
No inter-observer variability was assessed in this preliminary evaluation.
Not yet implemented or validated in real-time on embedded hardware platforms.
No inter-observer variability was assessed in this preliminary evaluation
Higher compression ratios (up to 38:1) come at the expense of signal quality

Implication

Supports efficient ECG data handling in wearables; leaves open prospective clinical validation of diagnostic fidelity.

Abstract

The growing adoption of wearable health monitoring and internet of things (IoT)-based medical devices has increased the demand for efficient electrocardiogram (ECG) compression techniques that reduce data size while preserving diagnostic fidelity. This paper proposes a compression method that combines linear predictive coding (LPC) with a residual thresholding mechanism. The ECG signal is modeled using a tenth-order LPC framework, and the residual is sparsified by retaining only the most significant components. The novelty of this work lies in applying threshold-based residual sparsification tailored to ECG signals, enabling high compression ratios (CR) while preserving clinically relevant waveform features. Experimental results using PhysioNet ECG data show that the proposed method achieves CRs of up to 38:1, while a moderate threshold (alpha is 0.005) provides a practical trade-off, achieving around 20:1 compression with preserved P wave, QRS complex, and T wave. The method offers a low-complexity solution suitable for wearable and IoT-based healthcare applications.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Hasan et al. (2026) studied Electrocardiogram (ECG) signal compression for cardiac monitoring. Linear predictive coding (LPC) with threshold-based residual sparsification (alpha=0.005) vs. LPC with exact (unthresholded) residual signal (alpha=0) was evaluated on Compression ratio (CR) and percent root-mean-square difference (PRD). Linear predictive coding with threshold-based residual sparsification achieved ECG compression ratios up to 38:1, and 20.07:1 (PRD 5.03%) at a moderate threshold while preserving waveform fidelity.

synapsesocial.com/papers/6ae28dedcbadf70daf6c3fb1 https://doi.org/10.11591/eei.v15i5.11688

View Full Paper

Ask AI

Bookmark

Share

Efficient electrocardiogram signal coding for medical devices using linear prediction

Key Points

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Efficient electrocardiogram signal coding for medical devices using linear prediction

Key Points

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading