Comparative Analysis of Oculography Methods For The Diagnosis Of The Vestibular System

There are various eye-tracking methods and the most common among them are methods of videooculography (VOG) and electrooculography (EOG). These methods are based on completely different physical principles, and accordingly, the signals received during the eye movements recording have different noise sources. Eventually, both methods register the same mechanical movement, and as respect, they should give consistent results. In this regard, the experimental study and analysis of accuracy of VOG and EOG while visual saccades registration performed using these methods of oculography simultaneously were carried out. 12 horizontal saccades to the left with an average amplitude of V-left = 150 +/- 50 deg/s and 10 saccades to the right with an average amplitude of V-right = 160 +/- 50 deg/s were recorded simultaneously by these two methods. The signals were processed by a low-pass filter with a cut-off frequency of 10 Hz and synchronized in time using crosscorrelation. Further, the saccades were averaged separately for each system and the left-right direction. The comparison of visual saccades was carried out using a linear regression model of the dependence of the eye velocity, recorded by EOG, on the eye velocity, recorded by VOG. A statistically significant good agreement of the eye velocities recorded using EOG and VOG during horizontal saccades was found. The obtained results of experimental studies have shown that despite the completely different physical principles these two methods based on and the presence of measurement errors of different nature, on average, the results of the eye movement registration by VOG and EOG systems are in excellent agreement with each other. The advantage of using EOG for measuring saccades and evaluating the vestibulo-ocular reflex (VOR)is shown.