Abstract
A new procedure is proposed for the automated detection of Focal Cortical Dysplasia (FCD) lesions on T1-weighted MRIs using volume-based discriminative features. Statistical features are obtained from of a set of neighboring voxels without using any computation that requires hard labeling of grey matter and white matter tissues. The significance of the proposed features is quantitatively evaluated with a Naive Bayes probabilistic approach, which is used for classification, and experiments are conducted on a total of 21 subjects with FCD lesions. The experimental results indicate that using the proposed features can achieve better detection rate and lower false positive rate for the FCD lesions compared to the widely used Antel's features.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
