Publication

Abstract : In biomedical sciences, the identification of cardiovascular diseases is more challenging and it is detected only after they become severe due to the delay in symptoms. On the other hand, experimentation with the cardiovascular system is unsafe and hence it is necessary to develop a cardiovascular system model which helps to predict the abnormalities in the early stage itself. In the existing arterial Windkessel model, the viscoelasticity characteristics are not considered, leading to error in the model and hence false prediction is unavoidable. To incorporate this characteristic, an improved fractional order Windkessel model is proposed. In this paper, fractional order is introduced in the existing 2-element, 3-element and 4-element Windkessel models. Further, a fmincon solver is used to optimize the model parameters and fractionality of the differential equation by minimizing the error between the clinical data and the model output. The simulation results indicate that the fractional order models provide less integral square error than the existing integer order Windkessel models. In specific, the fractional order 4-element Windkessel model provides better closeness to the clinical data than other fractional order models. This model is further used to study the behaviour of abnormalities (i) atherosclerosis and (ii) arterial stiffness.