Deep watermarking improves ECG signal security in telemedicine
ECGWmark: Deep Watermarking for Electrocardiogram Signals
Cryptography and SecurityComputer Vision and Pattern Recognition
Summary
Sending heart signals safely over the internet is tricky because small timing changes can mess up hidden data in the signal. The authors created a new deep learning method that hides information in ECG signals without losing important heart details, even if the timing shifts. Their method keeps medical details intact and can still read the hidden data accurately after signal changes. This could help protect patients’ information during remote heart checkups.
What this means in practice
- •For telemedicine platform developers: Embed patient metadata securely in ECG signals to maintain confidentiality and diagnostic quality during remote monitoring despite timing shifts.$Commercial implications: Enables commercial telemedicine providers to offer enhanced protected ECG transmission that preserves clinical accuracy and patient privacy.
- •For medical device engineers: Design ECG devices that embed and extract secure watermarks robustly against signal timing alterations without losing critical diagnostic features.
Tested on one dataset.
Authors
Kuan-Yu Chen, Kuan-Chen Wang, Yu Tsao, Jian-Jiun Ding
Abstract
Secure transmission of electrocardiogram (ECG) signals remains a major challenge in telecardiology. While traditional mathematical watermarking can embed patient metadata directly into ECG signals, it is highly fragile against temporal desynchronization such as circular phase shifts. In contrast, existing deep watermarking approaches for temporal signals, such as audio, typically rely on lossy quantization, which compromises the high-frequency micro-morphological details critical for clinical diagnosis. To bridge this gap, we propose ECGWmark, a fully convolutional and quantization-free deep watermarking framework for medical time-series. ECGWmark operates in a continuous, full-resolution latent space and utilizes a localized frame-level mechanism to improve the robustness to severe desynchronization. Crucially, we introduce a differentiable Physiological Loss guided by a clinical classifier to preserve diagnostic semantics. Experiments on the benchmark MIT-BIH database show that ECGWmark maintains channel-bit extraction accuracy above 90% under the evaluated circular temporal shifts. Furthermore, under a rigorous inter-patient diagnostic evaluation, Class-V sensitivity changes from 71.97% to 76.50%, whereas Macro-F1 changes from 35.25% to 34.71% and Class-S sensitivity from 8.89% to 1.39%; these results indicate mixed class-level effects.