The use of left ventricular assist devices (LVADs) in patients with advanced heart failure improves outcomes. The adaption of LVAD speed to individual needs is facilitated by the monitoring of pressures, volumes, and flows in the circulation. Particularly, the left ventricular volume (LVV) is a robust input variable for physiological LVAD control, and its monitoring reduces rehospitalizations. Bioimpedance measurements have been identified as a tool for assessing LVV. However, established techniques require elaborate in-vivo calibration or overestimate absolute LVV. This work provides methods and models for improved hemodynamic monitoring from bioimpedance measurements. Complex electric field distributions of bi- and tetrapolar measurement configurations within the aorta and left ventricle are translated into new techniques that estimate cardiac output (CO) and LVV. The fusion of bioimpedance and ultrasound sensors is realized to assess LVV requiring in-vitro calibration only. In-silico and in-vitro models are developed and applied to validate the developed methods. In-vitro phantoms mimic the dielectric and geometric properties of the left ventricle, mitigating costly animal testing. The optimal placement of electrodes to detect the center radius of the left ventricle and the flow through the aorta is determined in-silico for tetra- and bipolar configurations. A conclusive in-vivo experiment confirms the results obtained by the in-vitro and in-silico models, demonstrating that the proposed methods for the estimation of LVV outperform state of the art. The reliable estimation of LVV and CO has the potential to improve treatment in advanced heart failure and reducing mortality. The established methods lay the foundation for physiological control in LVAD therapy.weiterlesen