Informatique de gestion - Wirtschaftsinformatik

Epileptogenic lesion quantification in MRI using contralateral 3D texture comparisons

Jiménez del Toro, Oscar Alfonso ; Foncubierta-Rodríguez, Antonio ; Vargas Gomez, Maria-Isabel ; Müller, Henning ; Depeursinge, Adrien

In: Lecture Notes in Computer Science, 2013, vol. 8150, p. 353-360

Epilepsy is a disorder of the brain that can lead to acute crisis and if possible an intervention can help patients. The exact localization of epileptogenic lesions influences the outcome of epilepsy surgery. Magnetic resonance (MR) imaging is clinically used for lesion detection and treatment planning, mainly through simple visual analysis. However, visual inspection in MR imaging can be highly... More

Add to personal list
    Summary
    Epilepsy is a disorder of the brain that can lead to acute crisis and if possible an intervention can help patients. The exact localization of epileptogenic lesions influences the outcome of epilepsy surgery. Magnetic resonance (MR) imaging is clinically used for lesion detection and treatment planning, mainly through simple visual analysis. However, visual inspection in MR imaging can be highly subjective and subtle 3D structural abnormalities can be missclassified or not resected completely during the surgery. In this paper, a measurement of the differences in texture of the cerebral cortex between brain hemispheres is proposed. The 3D Riesz wavelet transform and the Hausdorff distance are used in MR T1 weighted images of the brain. A classification accuracy of 85% is obtained with 10 healthy control subjects and 8 patients with various types of epileptogenic lesions. The proposed quantitative measure of the texture alterations of these lesions allows an objective pre-surgical evaluation of patients undergoing epilepsy surgery.