Affecting an estimated 2.7 million Americans, epilepsy is one of the most widespread of neurological disorders. While advances in pharmacological agents have aided in the treatment and suppression of symptoms, a significant number, some 10- 20,000 individuals per year, continue to have uncontrolled seizures despite optimal medical management. Despite recent advances in modern imaging technology, neocortical lesions often remain difficult to localize due to their rapid spread throughout the cortex. Thus, there is an urgent need for the development of a method which is able to accurately localize these neocortical foci. Lack of such knowledge represents an important problem because without it, surgical intervention cannot be precisely tailored in patients where the source of ictal activity is often difficult to localize.;The long-term goal of the proposed research is to better understand seizure initiation and propagation throughout the cortex. Hence, it would be possible to use such information to tailor patient-specific surgical therapy. The rationale for the work presented in this dissertation is to develop new techniques which will aid in improved, potentially less invasive surgical therapies. This would translate into more limited, higher yield cortical resections or even ablation of specific seizure propagation pathways. The work presented in this dissertation utilizes time-series analysis tools based upon the concept of Granger Causality to identify and classify cortical networks in patients with neocortical-onset epilepsy. From the results of the studies performed in this work, it is observed that such analysis procedures are potentially useful in the identification of the epileptogenic generators in patients with intractable neocortical-onset epilepsy. Furthermore, the results suggest that the techniques developed in this thesis are predictive of post-surgical outcome and could be valuable in the pre-operative evaluation of these patients.
展开▼
