A major impediment to progress towards reproducible, rigorous preclinical research in epilepsy is that current experimental approaches often require prohibitively time and labor-intensive 24/7 video-EEG monitoring and inherently subjective scoring of seizures by human observers. In collaboration with the Datta lab at Harvard, who showed that complex animal behaviors are structured in stereotyped modules (“syllables”) at sub-second timescales arranged according to specific rules (“grammar”), we are developing a technique employing video imaging with a 3D camera and artificial intelligence (AI)-assisted video analysis to assess epileptic behavior in an automated and unbiased manner.
Instructor
Tilo Gschwind
Focusing on hippocampal network reorganization in temporal lobe epilepsy while tackling inherent problems of decade-old technology to advance epilepsy research, his project in the Soltesz lab provides an optimal opportunity to contribute to the interdisciplinary discourse between the fields of neuroscience and AI.
Email:gschwind@stanford.edu