Collaborating Investigator: Simon Warfield, Director of the Computational Radiology Laboratory

Working with Dr. Simon Warfield, the CIBC is developing computational tools to improve the localization of lesions in epileptic patients. Epilepsy is a chronic neurological disorder caused by disturbances in the normal electro-chemical function of the brain.  Epilepsy affects over 2.5 million Americans and has an estimated total annual health care cost close to $12.5 billion per year.  Medications for epilepsy often have significant side effects and currently medications fail to halt seizures in up to 20% of patients.  These patients are candidates for surgical intervention. Approximately 75 % of epilepsy patients have their first seizure in childhood.  Many become candidates for surgical intervention only after a long period of partially effective medication that can have debilitating educational and sociological side effects.

Many epileptic seizures are due to lesions.  Accurate identification and localization of lesions is critical to improved surgical intervention. Current state-of-the-art surgical interventions are limited by the risks and difficulty of identifying lesions prior to surgery.  Existing efforts to localize lesions rely upon scalp EEG, as well as the more accurate but invasive intra-dural EEG source localization, and the qualitative review of a range of imaging techniques such as MRI, MR spectroscopy, CT, PET, and fMRI. Technological innovations offer the prospect of improving foci localization and thereby dramatically improving patient outcomes.  The development of a highly accurate non-invasive localization of epileptogenic foci would also extend surgical intervention to patients previously deemed inoperable.  In current practice, the assessment of imaging data is carried out for each modality independently.  In this study, we seek to show that the interpretation of data from these imaging modalities benefits from the integration and simultaneous assessment of those images at the same time.

The overall goal of this project is to develop quantitative analysis algorithms to dramatically improve the capacity to detect and localize epileptogenic foci, in order to enable curative surgery for a significantly increased number of patients. The collaboration uses a number of CIBC software tools, including: Seg3D to segment different tissue classes from patient MRI scans, BioMesh3D to create patient-specific geometric models, and SCIRun/BioPSE to compute forward bioelectric field solutions and to estimate which dipole locations offer the best match to the patient EEG.  The CIBC’s visualization tools, ImageVis3D, SCIRun, and BioTensor, also help in exploring the patient specific geometry and bioelectric field solutions, to appreciate the relationship between the cortical anatomy and the seizure focus, and to communicate the results to the neurologists and neurosurgeons.