Improving seizure recognition by visual reinforcement

ObjectiveExact seizure description is an important diagnostic tool of epilepsy; however, epileptologists frequently get misleading data from lay eyewitnesses. The aim of this study is to find new methods for efficient training of seizure observation.MethodsTwelve video-recorded seizures were observed by four groups of subjects with different expertise level in epilepsy: (1) naïve observers, (2) pediatric residents, (3) epilepsynurses, and (4) experts in video-EEG monitoring. In half of the experiments, relevant parts of seizures were highlighted to direct the subjects’ gaze during observations. Eye motion data of the observers were recorded by a Tobii T120 Eye Tracking system. After watching a seizure, subjects were asked to fill out a questionnaire regarding semiology data. To reinforce associations between verbal and visual information, subjects watched again the same clip. Observation skill improvement was accessed by comparing questionnaire scores and eye fixation data.ResultsQuestionnaire mean scores increased and deviations decreased through the study demonstrating overall improvement for each group. Fixation time values were shorter in more experienced observers, however, this difference diminished by time reflecting the effect of the learning process. The chosen method for visual highlighting did not facilitate learning.ConclusionsOur pilot study – using eye-tracking assessment in clinical epileptology for the first time – shows that seizure observational skills can be significantly facilitated by a combined use of video examples and questionnaires. With the increasing popularity of eye-tracking methods in applied sciences, our results may lead to novel approaches in epilepsy education of both lay subjects and professionals.

► Exact seizure description is an important diagnostic tool of epilepsy.
► We looked for new methods for efficient training of seizure observation.
► Seizure observation was different among groups with different epilepsy experience.
► Our method improved seizure observation efficacy in all groups.