Differential Watermarking of Multilead ECG Baseline

Area of Science:

• Biomedical Engineering
• Digital Signal Processing
• Medical Informatics

Background:

• Digital watermarking is crucial for embedding auxiliary data in host records.
• Protecting sensitive patient data within medical signals is paramount.
• Existing watermarking methods may impact the diagnostic quality of physiological signals.

Purpose of the Study:

• To develop a novel digital watermarking method for electrocardiograms (ECGs).
• To embed auxiliary data within the ECG baseline without affecting diagnostic accuracy.
• To enhance the security and integrity of patient data in ECG records.

Main Methods:

• A new watermarking technique utilizing lead-to-lead differences in the ECG baseline.
• Baseline delineation followed by prediction of leads using Kirchhoff's voltage law or interpolation.
• Coding auxiliary data (patient information, measurement results) to mimic baseline noise.
• Replacing baseline noise with the watermarked data.

Main Results:

• The proposed method successfully embedded watermarks in 12-lead CSE ECG files.
• Achieved a watermark capacity of 3875 bits per second.
• Demonstrated that the diagnostic value of the ECG remains uncompromised after watermarking.

Conclusions:

• The novel ECG watermarking method effectively embeds auxiliary data.
• The technique preserves the diagnostic integrity of ECG signals.
• This approach offers a secure and efficient way to integrate patient data into ECG records.