Attention Estimation in Virtual Reality with EEG based Image Regression

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder affecting a certain amount of children and their way of living. A novel method to treat this disorder is to use Brain-Computer Interfaces (BCI) throughout the patient learns to self-regulate his symptoms by herself. In this context, researches have led to tools aiming to estimate the attention toward these interfaces. In parallel, the democratization of virtual reality (VR) headset, and the fact that it produces valid environments for several aspects: safe, flexible and ecologically valid have led to an increase of its use for BCI application. Another point is that Artificial Intelligence (AI) is more and more developed in different domain among which medical application. In this paper, we present an innovative method aiming to estimate attention from the measurement of physiological signals: Electroencephalogram (EEG), gaze direction and head movement. This framework is developed to assess attention in VR environments. We propose a novel approach for feature extraction and a dedicated Machine Learning model. The pilot study has been applied on a set of volunteer and our approach presents a lower error rate in comparison with the state of the art methods.