deepFCD: Multicenter Validated Detection of Focal Cortical Dysplasia using Deep Learning
We used clinically acquired 3-dimensional (3D) T1-weighted and 3D fluid-attenuated inversion recovery MRI of 148 patients (median age 23 years [range 2–55 years]; 47% female) with histologically verified FCD at 9 centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed MRI-negative in 51% of patients, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 cases with FCD (13±10 years). Applying the algorithm to 42 healthy controls and 89 controls with temporal lobe epilepsy disease tested specificity.
deepMask: Accurate Brain Segmentation in Malformations of Cortical Development
An in-house implementation of deep convolutional neural networks (V-Net) for brain extraction (removal of skull, dura mater, and cerebellum) using either T1-weighted alone or in conjunction with T2-weighted FLAIR MRI images in malformations of cortical development.
Plotly Dash application to generate textures maps from MRI using Advanced Normalization Tools (ANTsPy).
noelTexturesPygenerates 3D volume-based Relative Intensity and Gradient Magnitude maps derived from 3D T1-weighted MRI with isotropic resolution for computer-aided detection of focal cortical dysplasia (FCD).