Gaussian Mixture Model of Saccadic Eye Movements

This paper reports a study conducted to model saccadic eye movements based on a combination of Gaussian basis functions. Eye movement signal was recorded employing the electrooculography technique using a commercial bio amplifier that records the electrical activity of the eyes through surface electrodes. The Gaussian Mixture Model algorithm was employed for this purpose and implemented using MATLAB software. Modeling these eye movements is essential for feature extraction, processing, compression, transmission, and prediction applications. The proposed technique succeeded in modeling saccade based on root mean squared error, mean absolute error, mean percentage absolute error, and coefficient of determination, $\mathrm{R}^{2}$ , parameters employing 10 Gaussian basis components.