The development of a temporary cardiac pacing simulator: A training tool to enhance the management of post cardiac surgical patient care

Abstract

Temporary cardiac pacing (TP) is essential for managing haemodynamically unstable arrhythmias following cardiac surgery, yet its effectiveness depends on precise manual adjustments by clinicians. Despite its critical role, TP training remains inconsistent due to a lack of formal guidelines and inadequate simulation tools. Existing training methods fail to integrate key haemodynamic parameters and complex clinical scenarios, limiting their ability to fully prepare clinicians for real-world situations. This thesis presents the development of the Temporary Cardiac Pacing Simulator (TCPS), a novel training tool that bridges the gap between theoretical knowledge and hands-on experience. The TCPS incorporates multimodal physiological signals, realistic pacing modes, and advanced algorithms to simulate pacing failures and haemodynamic responses, providing real-time feedback and interactive learning. Additionally, the TCPS introduces a central venous pressure (CVP)-based approach to optimising atrioventricular (AV) delay, enhancing pacing efficiency and patient outcomes while exploring the feasibility of real-time AV delay optimisation in permanent pacemakers. Beyond the simulator, this research advances the techniques needed to further develop cardiovascular training tools. A GAN-based system (MC-WGAN) was developed to generate high-fidelity multimodal signals, addressing data scarcity and expanding training possibilities. Furthermore, advanced classification techniques, including ResNet architectures, were explored to improve automated multimodal and single-channel arrhythmia detection, enhancing the management of TP patients. Together, these contributions advance the field of TP devices, cardiovascular signal processing, and clinical training methodologies. By integrating novel simulation techniques, multimodal synthetic signal generation, and machine learning applications, this thesis provides a foundation for improved patient care, enhanced clinical education, and future developments in intelligent cardiac pacing systems.