Heart failure is the first cause of death in industrialized countries. Despite the use of medical therapy a certain percentage of patients (PTs) will progress to end stage heart failure. In this case the ventricular assist device (VAD) is a viable therapy that nowadays is used increasingly not only as a bridge to transplantation, but also as destination therapy and as a bridge to recovery. Therefore, there is a clinical need to understand how to estimate and modulate the optimal level of support (VAD rotational speed) to be provided according to PTs’ activity (resting, standing up, bicycle riding etc).This project is aimed at studying the effects of VAD speed modulation on PTs’ hemodynamics during physical activity (PA) and to compare them with a fixed VAD speed. To this aim clinical data will be acquired before and during VAD implantation surgery. Then two follow-ups will be organized to acquire data on VAD PTs performing different levels of PA at constant vs. manually increased VAD speed. Besides, due to the complexity of the issue involving both cardiovascular and respiratory variables, a computational cardio–respiratory simulator will be developed. It will be able to reproduce beat-to-beat PTs’ specific hemodynamic and respiratory condition (in terms of heart rate, flows, pressures, O2 and CO2 saturations, vascular resistances etc), at rest and during PA including the effects of VAD support.The present clinical study may contribute to improve VAD therapy with benefits on both PTs’ quality of life and on possible PTs’ improvements/recovery. Besides, the simulator will be a user friendly tool exploitable as a clinical support for VAD therapy optimization and PTs clinical management improvement. Moreover the simulator will be exploitable as a training tool for clinicians, VAD technicians, nurses etc. It will be also a computational test bench where different VAD speed modulation strategies (also from VAD manufacturers) can be simulated and tested.