Data hiding techniques for biomedical signals: A comprehensive review of steganography and watermarking approaches

Biomedical signals such as electrocardiography (ECG), electroencephalography (EEG), electromyography (EMG), and photoplethysmography (PPG) contain sensitive clinical information that requires robust protection during storage and transmission. Data hiding techniques, including steganography and watermarking, have been widely investigated to ensure confidentiality, integrity, and authentication while preserving the diagnostic quality of the original signal. This study presents a systematic literature review of biomedical signal data hiding research published between 2015 and 2025. Following the PRISMA methodology, 54 peer-reviewed journal articles were selected for detailed analysis. The reviewed studies were organized into a structured taxonomy comprising Classical and Advanced approaches. Classical methods, which include Time Domain, Transform Domain, and Hybrid Classical techniques, represent the dominant methodological group throughout the observed period. Advanced methods, encompassing Optimization-based, Intelligent, and Hybrid Advanced strategies, demonstrate a higher growth rate, reflecting an emerging shift toward more adaptive and learning-based data hiding solutions. Trend analysis of biomedical signal usage reveals that ECG is the most extensively investigated signal, followed by EEG, PPG, and EMG, suggesting a clear research growth hierarchy across signal modalities. Evaluation practices are primarily centered on imperceptibility, robustness, and capacity, which are most frequently assessed using Percentage Root-mean-square Difference (PRD), Bit Error Rate (BER), and payload-related metrics. Furthermore, security enhancement strategies integrating cryptographic techniques and blockchain-based frameworks are identified as promising directions for strengthening the protection of Patient Health Records (PHR). By consolidating existing taxonomies, trend findings, and evaluation strategies, this review highlights key research gaps, particularly the under exploration of non-ECG signal modalities and the limited adoption of intelligent methods, and provides a structured foundation to guide future developments in biomedical signal data hiding.