Department of Neurology and Brain Imaging Center, Montreal Neurological Hospital and Institute, McGill University, Montreal, Quebec, Canada
- Mots-clés : focal cortical dysplasia, epilepsy, magnetic resonance imaging, image processing, texture analysis
- Page(s) : 81-4
- Année de parution : 2003
In many patients, lesions of focal cortical dysplasia (FCD) may go unrecognized by standard radiological analysis. This is due to the fact that identification of many of these malformations on visual inspection of conventional MRI is difficult due to their subtlety and the complexity of the brain‘s convolutions. Quantitative MR image processing methods have the potential to help identify lesions that may be overlooked by conventional radiological evaluation. To increase the sensitivity of MRI for the detection of subtle lesions of FCD, we recently developed voxel‐based image post‐processing methods, including first‐order texture analysis and morphological processing modeled on known MRI features of FCD. Using these methods we were able to increase the sensitivity over conventional MRI analysis by more than 30%, while maintaining a high degree of reliability. The image processing methods we developed improve visual detection of FCD, even in cases where no lesion is obvious on MRI. Therefore, these techniques could allow a more precise evaluation of patients with partial epilepsy who could benefit from surgery.