Unique parameter identification of a cardiovascular system model using feedback control

Lumped parameter differential equation models are a common approach to modeling the cardiovascular system. However, there are highly non-linear valve dynamics inherent in these models which makes parameter identification difficult. Standard methods for parameter identification rely on gradient descent, which can often converge to wrong solutions, particularly as the number of parameters increases. This paper presents a new concept of parameter identification, applied to a 2 chamber model of the left ventricle systemic system. The changes in the parameters are treated as an actuation force into a feed back control system, where the reference output is taken to be steady state values of measured volume and pressure. The major advantage of the method is that when it converges, it must be at the global minimum, so that the correct solution is always found. The method is validated in both simulation and on a porcine model of pulmonary embolism. Very accurate matches to clinically measured left ventricle volume/pressure and aortic pressure waveforms are achieved, and the method gives considerable flexibility in capturing any required geometrical feature in the waveforms.