This research focuses on synthesizing physiological electrocardiogram (ECG) signals using a cardiac dipole model that mathematically represents the electrical activity of the heart. PR segments were used as the reference for signal normalization, and ...
This research focuses on synthesizing physiological electrocardiogram (ECG) signals using a cardiac dipole model that mathematically represents the electrical activity of the heart. PR segments were used as the reference for signal normalization, and a three-dimensional vectorcardiogram (VCG) model was designed based on the Frank Lead system. The synthesized 3D VCG data were utilized to generate Lead I and Lead II signals through inner product calculations, and the results were compared with actual measured data to evaluate the reliability and accuracy of the model. The analysis confirmed a high correlation between the synthesized Lead I and Lead II signals and the actual measured data, demonstrating that the proposed cardiac dipole model can physiologically reflect the electrical activity of the heart. Additionally, key ECG waveforms (P wave, QRS complex, T wave) were quantitatively analyzed using a cardioid-based mathematical modeling approach. The Integral Pulse Frequency Modulation (IPFM) model was applied to generate synthetic ECG signals that incorporate heart rate variability (HRV), resembling actual physiological characteristics. This research demonstrates that synthetic ECG signals can serve as valuable resources for overcoming limitations in clinical and research environments, such as privacy concerns, data imbalance, and noise. The proposed model suggests potential applications in various fields, including early diagnosis, treatment evaluation, and the development of automated algorithms for cardiovascular diseases. Future research will focus on extending the model to incorporate pathological characteristics and further validating its clinical applicability.