Automated Surfaced-based Detection of Focal Cortical Dysplasia using MR Fingerprinting

Focal cortical dysplasia (FCD) is a common pathology in medically intractable focal epilepsy and often difficult to detect by visual inspection of conventional MRI. We developed a framework for automatic FCD detection using surface-based processing of conventional MRI and MR fingerprinting data. Thirty-six patients with FCD and 48 healthy controls were included. Improved vertex-wise and cluster-wise performance was seen when MRF and FLAIR features were added to T1w data. A second-stage cluster-wise classifier showed efficacy to reduce false-positive clusters. Interim results of patient-level sensitivity of 76% and low false-positive clusters in controls supported potential clinical applicability of the proposed framework.