Human-Heart-Models for Formal Verification and Hardware-in-the-Loop Validation of Pacemakers

Pacemakers are an integral component of cardiovascular therapy. They are employed when a patient’s cardiac rhythm is insufficient to meet the individual demands. Given the considerable patient-specific variability over a day and throughout the lifespan, this long-term implant has to adapt to the individual patient’s changing needs and regulate the appropriate heart rate. In order to avoid potential risks to the patient, several factors have to be considered when pacing the heart. Stimulation at an inappropriate temporal phase can result in potential life-threatening adverse effects. Similarly, uncoordinated contraction among the heart chambers can lead to complications such as exercise intolerance, significantly impacting the patient’s quality of life. The growing complexity of pacemakers is driven by the need to accommodate the unique needs of patients. This has led to an increased reliance on pacemakers to perform their functions with greater autonomy, necessitating comprehensive testing to ensure their reliability in diverse scenarios. The complexity of this process is further compounded by the multitude of factors that affect the pacemaker. Consequently, it is essential to develop a comprehensive testing and verification environment that can reflect the diverse range of influencing factors and borderline cases the pacemaker may encounter over its lifetime.
In this thesis, a heart model is presented that is capable of mirroring the Electrical Conduction System of the Heart (ECSH), the chambers and the valves of the heart. The model is capable of including the status of the chambers and valves as a function of the stimulation interval and the Action Potential Duration (APD). This enables the investigation of diverse scenarios, including pacing in the vulnerable phase, the coordination of the chambers, the number of contractions for various diseases concerning the ECSH and the heart chambers.
The model is demonstrated to be capable of modeling 14 diseases concerning the ECSH, the generation of pulses within the heart and the heart chambers, such as various atrioventricular blockages, bradycardia, premature contractions, flutter, and variances in the QT interval due to the menstrual cycle. The objective is to demonstrate the value of formal verification and Hardware-in-the-Loop (HIL) techniques for the development and improvement of increasingly complex medical devices in the future.
The proposed heart model is transformed into a formal verification model and a HIL model.
The latter is then evaluated in interaction with a real pacemaker. A formal verification model of a pacemaker is constructed to assess the interaction between the formal verification model of the heart and a pacemaker. A particular challenge is to incorporate the various impact factors while keeping the model executable, especially since it is essential that the heart model interacts with the pacemaker in real time for HIL testing.
The HIL tests indicate the potential for life-threatening conditions concerning the number of contractions per minute, pacing during the vulnerable phase, and the coordination of the chambers of the heart. The results also show the impact on the patient’s safety of different diseases, of the chamber monitored by the pacemaker, of the frequency of sinoatrial node (SA node) self-stimulation and of the replacement rhythms. Our pacemaker model for formal verification shows promising results for the double chamber pacemaker. The single chamber model, however, is not yet suitable for all diseases. To the best of our knowledge, the pacing during the vulnerable phase and the chamber coordination have not been evaluated before.