Kernel Fisher for Shape Based Classification in Epilepsy

  • Authors:
  • N. Vohra;Baba C. Vemuri;Anand Rangarajan;R. L. Gilmore;S. N. Roper;C. M. Leonard

  • Affiliations:
  • -;-;-;-;-;-

  • Venue:
  • MICCAI '02 Proceedings of the 5th International Conference on Medical Image Computing and Computer-Assisted Intervention-Part II
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present the application of Kernel Fisher in the statistical analysis of shape deformations that might indicate the hemispheric location of an epileptic focus. The scans of two classes of patients with epilepsy, those with a right and those with a left medial temporal lobe focus (RATL and LATL), as validated by clinical consensus and subsequent surgery, were compared to a set of age and sex matched healthy volunteers using both volume and shape based features. Shape based features are derived from the displacement field between the left and right hippocampii of a healthy subject/patient. The results show a significant improvement in distinguishing between the controls and the rest (RATL and LATL) using only the shape as opposed to volume based features. We also achieve a reasonable improvement in the efficiency to distinguish between RATL and LATL based on shape in comparison to volume information. It should be noted that automated identification of hemispherical foci of epilepsy has not been previously reported.