Add
Correspondence should be addressed to:
Sara K. Inati
NINDS, National Institutes of Health
Building 10, Room 7-5680
10 Center Drive Bethesda, MD 20892-1445
Office: (301) 435-6269
Keywords: Structural MRI, Machine learning, Epilepsy, Focal cortical dysplasia
Number of Figures: 7
Number of words (approximate): TBD
Funding: This work was supported by the Intramural Research Program of the National Institute of Neurological Disorders and Stroke.
Declarations of Interest: None
Introduction
Drug-resistant focal epilepsy is a disabling disease, leading to increased morbidity and mortality as well as psychosocial and economic impairment \citep{Lawn2004, McCagh2009, Mohanraj2006, Theodore2006}. In appropriately selected patients, surgery is a proven treatment, with long-term seizure freedom achieved in approximately 65% and 45% of patients with temporal and extratemporal lobe epilepsy, respectively (Tellez-Zenteno 2005). These seizure outcomes are significantly improved in patients with identifiable epileptogenic lesions on MRI (Tellez-Zenteno 2010). Some epileptogenic lesions easy to detect. In other cases, they can be subtle or equivocal, and no appreciable lesions are identified in 15-30% of patients (Duncan 2016). It is estimated that approximately half of these non-lesional patients will be found to have focal cortical dysplasia by pathology (Chapman 2005, Lee 2005, McGonigal 2007 or Bien 2009, Lerner 2009); conversely 50-80% of patients with FCD have normal-appearing MRIs (Besson 2008). Therefore, improved identification of subtle focal cortical dysplasias (FCDs) is of great clinical importance as it may improve surgical outcomes in these patients.