358 An Anatomical Study of the Sylvian Fissure Using a Large Number of MRI Data Sets

Abstract Introduction Accurate manoeuvring of the Sylvian fissure borders to provide an optimal operating corridor is vital for minimising retraction. The anatomy can pose technical challenges impacting overall success procedure. Yaşargil previously proposed a classification distinguishing four anatomical variants correlating with surgical difficulty. We aim utilise open-source Human Connectome Project MRI data set computational based on morphology. Method MR images 2226 hemispheres were included in this study. Morphometrical parameters designed and refined MATLAB simulations. Combinations analysed using K-Means Clustering. Results Following segmentation removal outliers, 1693 fissures remain analysis. Clustering revealed three clusters sulcal bulk displacement measure (deviation towards frontal or temporal lobe) distance (width fissure) (silhouette score = 0.383). Conclusions Analysis large neuroimaging cohort suggested existence weakly separated variants. propose that looking at quantitative scores rather than class allocation would be more informative preoperative planning.