Biomedical Engineering - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemEpileptic seizures: mechanisms and forecasting( 2018)Seizure forecasting, like weather forecasting, was once considered the domain of charlatans and purveyors of science fiction. However, neuroscience has now advanced to the point of translating seizure forecasting research into widely available clinical applications. Just like weather apps that report the probability of rain on a given day, it is now conceivable that devices will inform people with epilepsy about their current likelihood of having a seizure. This information could be life-changing: restoring a sense of control and the ability to participate in everyday activities. Over 65 million people around the world have epilepsy; one third cannot control their seizures with medication. The unpredictability of seizures can be devastating, leading to persistent anxiety, exclusion from day-to-day life, serious injury or death. The aim of this thesis is to develop a clinically useful framework for forecasting seizures. The presented research addresses several key questions towards this goal: What drives seizure transitions? Are there underlying rhythms governing seizure onset? If underlying rhythms exist, how can they be integrated into a single determination of an individual's seizure likelihood? By presenting answers to these questions this thesis aims to form the basis for an innovative approach to seizure forecasting.
-
ItemTracking neural dynamics( 2014)Epilepsy affects around 1% of the world's population, of which roughly a third are refractory to treatment. A major impediment to the development of effective treatments for these patients is the lack of understanding of the mechanisms underlying the disorder. In this thesis, a model-based approach is developed to provide some insight into these mechanisms. Initially, it is demonstrated that parameters linked to physiology from nonlinear models of the brain can be estimated accurately under various conditions. An animal model of epilepsy is used to illustrate that this computational model of the brain can be used to demonstrate variability in seizure mechanisms between animals, and over time. In particular, it is demonstrated that seizures group across days in the animal model, and that for each group, different physiological mechanisms are involved. This provides evidence that in the animal model studied seizures evolve, and that this evolution is linked to seizure grouping. With the knowledge that seizures are subject-specific and evolve over time, it may be possible to develop techniques that use control system theory to continuously alter therapy based on estimated physiological parameters inferred from recorded EEG, to provide treatments that are more efficacious. odel of epilepsy is used to illustrate that this computational model of the brain can be used to demonstrate variability in seizure mechanisms between animals, and over time. In particular, it is demonstrated that seizures group across days in the animal model, and that for each group, different physiological mechanisms are involved. This provides evidence that in the animal model studied seizures evolve, and that this evolution is linked to seizure grouping. With the knowledge that seizures are subject-specific and evolve over time, it may be possible to develop techniques that use control system theory to continuously alter therapy based on estimated physiological parameters inferred from recorded EEG, to provide treatments that are more efficacious.