Validation of automatic MRI hippocampal subfield segmentation by histopathological evaluation in patients with temporal lobe epilepsy

ObjectiveThe present study validates the results of automated hippocampal subfield segmentation with histopathology in epilepsy patients undergoing surgery.MethodsWe performed an on presurgical three-dimensional, T1-weighted magnetization Prepared Rapid Acquisition Gradient Echoes Magnetic Resonance Imaging (MRI) data 25 unilateral mesial temporal lobe due to sclerosis (HS), using Freesurfer Version 6.0. The resulting volumes cornu ammonis (CA) subfields CA1, CA2/3, CA4 and dentate gyrus (DG) were compared histopathological cell count.ResultsWe found a significant correlation between subregion CA2 CA1 (p = 0.0062), CA2/3 0.004), 0.0062) DG 0.0054), CA3 0.0132), 0.0004), 0.0032) 0.0037), as well 0.0115), < 0.0001), 0.0001) 0.0001). finding HS type 1 could correctly be classified all cases MRI.SignificanceThe shows correlations evaluation hippocampus, thereby validating method. As differential involvement different may associated clinical parameters outcome after surgery, is also promising for prognostic purposes.