Abstracts

Rationale: Hippocampal(HP) and its related structural(amygdala;AMG, entorhinal cortex;EC, thalamus;TH) volumetry is a widely used clinical tool for the detection and lateralization of mTLE. Traditionally, the volume of these structures is measured by manually segmenting on serial sections of a T1-weighted MRI. The manual approach requires a trained operator in a reliable and consistent manner. Automated software-based segmentation now allow us to use MRI to obtain estimates of these structural volume without requiring manual input. However, few of these studies explicitly compared automated volume estimates with the corresponding manual estimate in the same subject. In the studies in which a direct comparison of manual and automated techniques was made, an automated segmentation technique was used Freesurfer. The aim of this study is to determine if automated volumetric methods are suitable for detection of hippocampal and its adjacent regional atrophy in a mTLE cohort with unilateral hippocampal sclerosis. The outcomes of this study will help determine whether automated techniques should replace the manual technique as the standard volumetric methodology for the detection and lateralization of hippocampal atrophy in a clinical setting. Methods: Volumetric MRIs were measured on 26 patients with mTLE (16 left, 10 right) and 20 age- and gender-matched healthy controls. Volumes of hippocampus, subcortical regions (thalamus, amygdala) and entorhinal cortex were measured manually, and using the automated reconstruction software (FreeSurfer). The sensitivities of two techniques were compared by determining the effect size for average volume reduction in patients with mTLE compared to controls. Results: Significant volume reduction in affected hippocampi in mTLE compared to controls was detected by manual hippocampal volume measurement (p<1.0x10-6, effect size 16.0%), FreeSurfer (p<0.05, effect size 5.5%) after correction forbrain volume. FreeSurfer correlated reasonably with this manual segmentation on pathological HP in left (r=0.69, p<0.01) and right (r=0.63, p<0.01) mTLE patients. Contrary to previous reports, the spatial correlation between manual and automated segmentation was reduced in affected hippocampi, suggesting the accuracy of automated segmentation was increased in pathologic brains. The discriminant function analysis revealed that patients with mTLE were best distinguished from one another using a combination of its hippocampus and amygdala volumes by manual volumetry, but not by FreeSurfer. Conclusions: Expert manual volumetry is more sensitive than automated methods for the detection of hippocampal atrophy associated with mTLE. Volumetric data obtained with automated segmentation of HP and AMG approximate data from manual tracings. Our data suggest that automated segmentation may suggest to be clinically useful for evaluating the nature and extent of structural damage in patients with mTLE and may increase diagnostic classification of patients. Further studies with more patients should be warranted.