Mathematical Modeling of the Human Smooth Pursuit System

This licentiate thesis concerns mathematical modeling and identification of the the human smooth pursuit system (SPS) and the application of the models to motor symptom quantification in Parkinson’s disease (PD). The SPS is a complex neuromuscular system governing smooth pursuit eye movements (SPEM), and the task is to keep a moving target in the visual field. Diagnosing and quantifying the disease is done by interview and clinical observation which requires hours of interaction between the patient and a qualified clinician. Acquiring a better understanding of the SPS cast in mathematical models may be a first step towards developing a technology that allows for fast and automatic PD staging. Lately, the increased performance and accessibility of eye tracking technologies have generated a great deal of interest in the commercial sector. This thesis presents an effort towards developing more sophisticated data analysis techniques in an attempt to extract previously hidden information from the eye tracking data and to open up for new more advanced applications. The SPS relates gaze direction to visual stimuli and may thus be viewed as a dynamical system with an input and an output signal. This thesis considers various parametric and non-parametric blackand grey-box models, both linear and nonlinear, to portray the SPS. The models are evaluated to characterize the SPS in different individuals and to look for discrepancies between the SPS function of healthy controls and Parkinson patients. It is shown that disease does indeed impair the system and that the effects are distinguishable from those of healthy aging.