Abstracts

Rationale: To apply automated quantitative volumetric MRI analyses to patients diagnosed with Rasmussen's encephalitis, determine the predictive value of hemispheric and lobar volumetric ratio measures, and to assess regional atrophy difference and monitor disease progression using these measures.Methods: Volumetric T1-weighted MPRAGE sequences were analyzed with Brainsuite, on a cohort of 19 patients with diagnosed Rasmussen's encephalitis. Nine patients had multiple MRIs for a total of 42 studies. An age and gender matched control cohort of 42 normal subjects was used for comparison. Standard procedures of the freely available software Brainsuite were used to produce the volume of a large set of bilateral segmented cortical and deep brain regions, which were then used to generate ratios of the affected hemisphere divided by the unaffected hemisphere. These ratios were then used as input to a logistic regression classifier, which was trained to discriminate the patient from normal controls. These ratios were also used to assess regional atrophy difference. Correlation of volumetric ratios with epilepsy duration was examined as well.Results: Our method produces clearly defined separation of Rasmussen's patients from normal control subjects based on hemispheric and frontal ratio of brain volume, both with prediction accuracy of 0.95. The performance of classifier using hemispheric and frontal ratio both attains a sensitivity of 0.93 and a specificity of 0.98. Based on the HRvol classifier it was determined that a hemispheric ratio of less than 88.56% gives 95% probability of a patient belonging to RE group. Insula showed significantly more severe atrophy compared to all the other cortical regions. The 9 patients with multiple scans over time showed progressive atrophy of the affected hemisphere. Atrophy of the frontal lobe and insula correlated significantly with epilepsy duration.Conclusions: The automated quantitative volumetric analysis presented here provides accurate volumetric assessment of the MRI for patients with Rasmussen's encephalitis. The volumetric analysis also provides objective assessment of disease progression, and could be used as a complimentary tool for all patients with Rasmussen's encephalitis for their routine followup. This research was supported by NIH Grant R01-NS074980. Normal control data collection and sharing for this project were funded by the Pediatric Imaging, Neurocognition and Genetics Study (PING) via the National Institute of Health Grant RC2DA029475 and the National Institute on Drug Abuse and the Eunice Kennedy Shriver National Institute of Child Health and Human Development.