Identifying lesions in paediatric epilepsy using morphometric and textural analysis of magnetic resonance images

We develop an image processing pipeline on Magnetic Resonance Imaging (MRI) sequences to identify features of Focal Cortical Dysplasia (FCD) in patients with MRIvisible FCD. We aim to use a computer-aided diagnosis system to identify epileptogenic lesions with a combination of established morphometric features and textural analysis using Gray-Level Co-occurrence Matrices (GLCM) on MRI sequences. The model will be validated on paediatric subjects. Preliminary morphometric analysis explored the use of computational models of established MRI features of FCD in aiding identification of subtle FCD on MRI-positive subjects. Following this, classification techniques were considered. The 2-Step Naive Bayes classifier was found to produce 100% subjectwise specificity and 94% subjectwise sensitivity (with 75% lesional specificity, 63% lesional sensitivity). Thus it correctly rejected 13/13 healthy subjects and colocalized lesions in 29/31 of the FCD cases with MRI visible lesions, with 63% coverage of the complete extent of the lesion using supplied lesional labels.

Name thesis:degree_name

Master of Science (MSc)

Discipline thesis:degree_discipline

Modelling and Computational Science

Grantor

University of Ontario Institute of Technology

Year dc:date.issued

2017

Author dc:creator

• Kulaseharan, Sancgeetha

Advisors dc:contributor.advisor

• Ebrahimi, Mehran
• Widjaja, Elysa

• Computer-aided diagnosis
• Epilepsy
• Focal cortical dysplasia
• Morphometric
• Textural analysis

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10155/877

OAI identifier oai:identifier

oai:ontariotechu.scholaris.ca:10155/877