Evaluation of an MRI-compatible mock circulation loop for a total artificial heart using lumped parameter modeling

Sammanfattning

Heart failure is a growing health challenge with limited treatment options, where total artificial hearts (TAHs) can offer one potential solution to the current shortage of heart donors by replacing the heart’s function. To evaluate the function of TAHs without implantation in a body, mock circulation loops (MCLs) can be used, but their ability to accurately replicate human physiological conditions remains uncertain. In this thesis, the aim was to evaluate an MRI-compatible mock circulation loop (MRIMCL) and its potential to provide physiological conditions for testing the function of a total artificial heart, as well as to identify any undesired or non-physiological influences of the MRI-MCL on the TAH’s performance. To achieve this, measurements of the outlet pressure and flow from the TAH connected to the MRI-MCL were collected. The data was used to tune and validate lumped parameter models (LPMs) of the MRI-MCL to better understand its behavior and key components. An established physiological LPM was modified to include the TAH, thus establishing a physiological baseline for the TAH’s response in the body. The developed MRI-MCL models were consequently tuned to the measured data from the MRI-MCL and the TAH-in-body model to find the optimal system settings for the MRI-MCL to replicate physiological conditions. The results indicate that the MRI-MCL appears to influence the hemodynamic function of the TAH in both physiological and non-physiological ways. The developed models, while simple in design, seem to capture the overall shape of the MRI-MCL measurements, but miss finer details. Good following of the physiological model highlights the limitation of simulating TAH behavior in the body using LPMs. A higher-complexity model showed a better fit to the measured data from the MRI-MCL, indicating potential for better capturing using such models. In conclusion, the MRI-MCL appears to replicate key physiological features, such as peripheral resistance and arterial compliance. However, its limitations at the extremes suggest a need for design improvements to achieve higher physiological replication. These findings also highlight the challenge of designing MRI-compatible systems that maintain physiological accuracy, advocating for improved design practices combined with advanced modeling techniques. Such improvements can support the continued development and evaluation of TAHs under realistic physiological conditions, without relying on in vivo testing.