AZTDIS--a two-phase real-time ECG data compressor

Area of Science:

• Biomedical Engineering
• Signal Processing
• Data Compression

Background:

• High sampling rates in electrocardiogram (ECG) signal acquisition generate substantial redundant data, posing challenges for storage and transmission.
• Efficient data compression is crucial for managing large ECG datasets while preserving diagnostic fidelity.

Purpose of the Study:

• To introduce and evaluate AZTDIS, a novel real-time ECG data-compression algorithm.
• To compare the performance of AZTDIS against the established AZTEC algorithm in terms of data reduction and signal fidelity.

Main Methods:

• AZTDIS identifies significant ECG signal samples and encodes linear segments between them using linear interpolation.
• The algorithm adaptively selects significant samples based on user-defined maximal error (epsilon), prioritizing accuracy in high-activity regions.
• Performance was evaluated by comparing root-mean-square error (RMSerr) at equivalent bit rates with the AZTEC algorithm.

Main Results:

• AZTDIS achieves a considerable improvement in RMSerr compared to AZTEC at the same bit rate.
• An average RMSerr of 9.715 was obtained with AZTDIS at a low average bit rate of 0.543 bits per sample.
• When tuned to a similar bit rate as AZTEC, AZTDIS yielded an average RMSerr of 5.554, significantly lower than AZTEC's 19.368.

Conclusions:

• The AZTDIS algorithm offers superior ECG data compression efficiency and accuracy compared to AZTEC.
• AZTDIS effectively balances data reduction with the preservation of clinically relevant ECG signal information.
• This algorithm presents a promising solution for efficient storage and transmission of high-fidelity ECG data.