Trend prediction technique for Computer Visual Syndrome (CVS) with eye tracking as support

This research presents information about CVS, a syndrome that affects visual health after spending a long time in front of a computer; from a list of its symptoms, it was possible to determine some detectable patterns among the participants, showing the possible development of this syndrome. We looked for a way to implement eye tracking as a support tool capable of collecting information about a person’s vision, such as: eye fixation points, fixation time, as well as blinking or impulsive eye movements. The main objective of the data obtained is to train a neural network model, capable of detecting the most common patterns of a person; by classifying these patterns, it is sought to predict whether this person has a high probability of acquiring the syndrome or not. Finally, we obtained an accuracy of 97.5%.