Embedding and retrieving private metadata in electrocardiograms

Area of Science:

• Biomedical Engineering
• Data Security
• Signal Processing

Background:

• Increasing identity theft necessitates robust protection for sensitive medical data.
• Current methods like encryption have limitations; alternative approaches for securing patient information are needed.
• Medical time-series signals, specifically Electrocardiograms (ECG), contain vital patient information requiring secure handling.

Purpose of the Study:

• To develop a novel method for embedding and retrieving sensitive patient metadata within ECG signals.
• To enhance medical data security by fusing sensitive information directly into the signal.
• To provide data authentication and origin assurance for medical signals.

Main Methods:

• Utilizing watermarking principles to embed sensitive numerical data (e.g., social security number, birth date) into ECG signals.
• Implementing multistage watermarking with both robust and fragile components for resilience and fault tolerance.
• Developing techniques for efficient embedding and retrieval of the hidden metadata without distorting diagnostic ECG characteristics.

Main Results:

• Successfully embedded and retrieved sensitive patient data within real ECG signals.
• Demonstrated that the watermarking technique does not negatively impact crucial ECG characteristics for medical diagnosis.
• Validated the resilience and fault tolerance of the multistage watermarking approach.
• Showcased the potential for data ownership authentication and origin assurance.

Conclusions:

• The proposed watermarking scheme is a viable method for securely embedding sensitive metadata into ECG signals.
• This technique offers a promising alternative to traditional encryption for protecting patient privacy in medical data.
• The method ensures data integrity and authenticity while maintaining the diagnostic utility of ECG signals.