Focal cortical dysplasia (FCD) is a surgically-amenable epileptogenic developmental malformation. On MRI, FCD typically presents with cortical thickening, hyperintensity, and blurring of the gray-white matter interface. These changes may be visible to the naked eye, or subtle and be easily overlooked. Despite advances in MRI analytics, current surface-based algorithms do not detect FCD in >50% of FCD lesions.

We propose a novel algorithm to distinguish FCD from healthy tissue on MRI voxels. Our method harnesses feature learning capability of convolutional neural networks (CNN), a powerful deep learning paradigm. The algorithm was trained and tested on data from the Montreal Neurological Institute (MNI) and tested on independent data from MNI and eight sites worldwide.