Advancing Cardiovascular Monitoring: Translation and Restoration of Photoplethysmogram using Super Operational Neural Networks in 1D-CycleGAN Framework

Cardiovascular diseases endure as a predominant cause of death on a global scale, requiring precise monitoring methodologies. The current study confronts the issue of limited accuracy in estimating cardiovascular parameters through the introduction of a methodology that exploits Photoplethysmogram (PPG) signals. By employing a Translation Through Restoration GAN (TTR-GAN) framework in conjunction with Super Operational Neural Networks (Super-ONNs), the system adeptly translates and restores PPG signals from the wrist to the finger. Through extensive experimentation, a substantial 15% enhancement in accuracy is evident when compared to traditional methods such as logistic regression and support vector machines. Furthermore, the proposed system excels in terms of precision, recall, and the area under the receiver operating characteristic curve (AUC). This study presents a methodical approach to augmenting accuracy in the estimation of cardiovascular parameters, offering the potential for enhanced healthcare protocols, early complication identification, and improved patient care outcomes.