Use of the Extended Kalman Filter for biological models accuracy

Abstract

The Extended Kalman Filter (EKF) has been shown to provide accurate estimation of states in complex nonlinear systems. However, the application of the EKF in biological models is still scarce. In order to demonstrate how an EKF can be used for measurement accuracy in biological models, a simplified circuit model of the retina is designed. Our model is comprised of diodes, linear electrical components and dependent voltage sources. The model is used to validate the use of the EKF for accurate measurement of a biological model's states under noisy conditions. An EKF and standard Kalman Filter (KF) are designed in Matlab and compared for voltage estimation accuracy at each cell layer of the retina model presented. Our results demonstrate that the EKF has better estimation capabilities in noisy conditions then the KF. Therefore an EKF can be utilized in biological models to improve measurement accuracy.