We
provide a dataset that includes visualizations of eye-tracking scanpaths with a
particular focus Autism Spectrum Disorder (ASD). The key idea is to transform
the dynamics of eye motion into visual patterns, and hence diagnosis-related
tasks could be approached using image analysis techniques. The image dataset is
publicly available to be used by other studies aiming to experiment the
usability of eye-tracking within the ASD context. It is believed that the
dataset can allow for the development of further interesting applications using
Machine Learning or image processing techniques. For more info, please refer to
the publication below and the project website.
Original Publication:
Carette, R., Elbattah, M., Dequen, G., Guérin, J, & Cilia, F. (2019, February). Learning to predict autism spectrum disorder based on the visual patterns of eye-tracking scanpaths. In Proceedings of the 12th International Conference on Health Informatics (HEALTHINF 2019).