A Secure and Efficient Framework for ECG Signal Digitization and Encryption for Internet of Medical Things Applications.

The Internet of Medical Things (IoMT) is transforming healthcare through smart devices, real-time monitoring, and cloud-based data exchange. Electrocardiogram (ECG) signals are crucial in diagnosing cardiovascular conditions; however, their sensitivity is still a major challenge for secure transmission and storage. This paper presents a novel content-based encryption framework tailored for ECG data. Unlike conventional approaches that process either raw signals or images separately, the proposed method first extracts accurately ECG waveforms from 2D images while handling noisy or overlapping traces through a dedicated preprocessing and segmentation pipeline. A two-layer chaotic scheme is then employed to encrypt the extracted 1D signals: the Arnold Cat map (ACM) for spatial scrambling and Double Random Phase Encoding for frequency-domain transformation. These cryptographic modules can also operate directly on native digital ECG signals and integrate seamlessly with paper-based ECG datasets, ensuring broad applicability across acquisition formats. Experimental results on a 12-lead ECG dataset demonstrate high digitization accuracy (Structural Similarity Index = 0.94), fast execution (0.0036 s), strong encryption (entropy = 7.3, correlation = 0.0006), and perfect signal recovery (PSNR = ∞). Overall, the framework provides a secure and efficient solution for IoMT-driven healthcare systems, outperforming existing ECG encryption methods across standard evaluation metrics.