- Our normative modeling approach is based on local image filters used in computer vision and allows for estimates of outlierness and similarity.
- Focal cortical dysplasias as well as heterotypic areas of normal cortex appeared as outliers in our model.  FCDs appeared similar in our features to some of these regions, particularly the anterior insula and mesial temporal paralimbic cortex.
- We automatically detected FCDs with a sensitivity of 80% and specificity of 70% using a small training set. Automated FCD detection is of great value in presurgical evaluation of patients with drug resistant epilepsy.
Number of Figures: 4
Number of Tables: 2
Number of words (approximate): 3999
Number of references: 34

Introduction

Focal cortical dysplasia (FCD) is a relatively common cause of drug-resistant focal epilepsy, and particularly of MRI-negative (MRI-) epilepsy.  Approximately 15-25% of patients being evaluated for epilepsy surgery have apparently normal MRIs; FCD is diagnosed pathologically in 25-50%  of these patients \cite{Blumcke2017,Lerner2009,Bien2009}.  Because post-operative seizure outcomes are significantly worse in MRI- compared to MRI-positive patients \cite{Téllez-Zenteno2010}, improved identification of subtle FCDs is of great clinical importance as it may lead to better surgical outcomes.
FCDs are variable in their histopathological and radiographic appearances.  Pathological abnormalities range from cortical dyslamination in type I, to the additional presence of large dysmorphic neurons in type IIa (FCDIIa), and balloon cells and more prominent dysmyelination in type IIb (FCDIIb) \cite{Blümcke2011}.  Typical radiographic findings include changes in cortical thickness, image intensity (including the transmantle sign in FCD type IIb), gray-white junction (GWJ) blurring, and atypical cortical folding patterns \cite{Adler2017a,Kini2016}.  These findings, however, exist on a spectrum, with features appearing inconsistently, even within and across FCDIIb lesions \cite{Hong2017}.  They can also be extremely subtle, with up to 80% of small bottom-of-sulcus dysplasias being missed on routine visual inspection \cite{Besson2008}.
A number of post-processing methods have therefore been developed to aid in FCD detection.  These range from creation of synthetic contrasts to highlight areas of interest (such as the Morphometric Analysis Program (MAP) \cite{Huppertz2005}) to machine-learning based fully automated detection methods  \cite{Adler2017,Ahmed2015,Hong2014}.  Across these methods, two key challenges have been 1) selection of optimal features to describe FCD lesions, and 2) accounting for variability within and across FCD lesions, as well as in normal cortex.  While a wide variety of features have been investigated, most methods have used features derived from either voxel-based morphometry \cite{Huppertz2005,Martin2017} or surface-based morphometry (SBM) (see review in \cite{Kini2016}).  To account for normal and pathological variability, large training sets are typically used, with each new patient's data compared to a normative database; performance tends to be dependent on the size of both the lesional and normative training sets \cite{Jin2018}.  
In this work, we implemented a novel set of 3D multiscale local image features across multiple MR contrasts.  Similar features have been previously shown to efficiently represent the local statistics of natural images \cite{Simoncelli2001}.  We show that our feature set encompasses information contained in several measures previously used for FCD detection.  We then use a normative modeling approach to characterize normal cortical variability.  In this representation, we were able to detect FCDs as outliers in a relatively circumscribed portion of the feature space, but only after normalizing for the expected appearance of normal cortex in healthy volunteers at each location.  Local normalization corrected for the expected atypical appearance of some normal cortical regions such as the anterior insula, which demonstrated similar underlying features to our FCDs; it also corrected for the initial similarity of our FCD lesions to the expected appearance of normal cortex at their given location.

Materials and Methods

Study Participants

From our surgical epilepsy imaging database, we retrospectively identified 15 consecutive patients undergoing presurgical evaluation for drug-resistant focal epilepsy from 2014-2019 with: 1) radiologically apparent (MRI+) or histologically proven (MRI+ or MRI-) FCDs; and 2) our standard 3T MRI epilepsy structural imaging protocol.  Patients were excluded if they underwent other MR imaging protocols or had low image quality on visual inspection. The control group consisted of 30 healthy volunteers (HVs) scanned using the same imaging protocol with no previous history of neurologic, psychiatric, or other significant medical illnesses that may affect the central nervous system. Data were collected at the National Institutes of Health (NIH) Clinical Center (Bethesda, MD). All participants were enrolled in an Institutional Review Board-approved research protocol; informed consent was obtained from all participants.

Lesion Labels

For MRI+ patients, lesions were traced in the volume using the Analysis of Functional NeuroImages (AFNI) software package \cite{Cox1996a} by an experienced neurologist using the T1, informed by the T2 and FLAIR images when necessary. For MRI- patients, the postoperative T1 was registered to the preoperative T1 in the same manner as described below for T2 and FLAIR images; the resected region was manually traced using AFNI. Lesion masks were mapped onto the smoothed white matter surface using AFNI's 3dVol2Surf function.

MRI Acquisition Protocol

All participants were scanned on a Philips Achieva 3T MRI scanner in the NIH Clinical Center Radiology Department as follows: 1) 3D T1 weighted MPRAGE (T1): TR = 6.8–7.2, TE = 3.2 ms, TI = 900ms, flip angle = 90, voxel size = 0.75 x 0.75 x 0.8, acceleration factor 2 in slice direction, acquisition time = 7:02 min; 2) 3D T2 weighted FSE (T2): TR = 2500, TE = 225–245, voxel size = 1 x 1 x 2 or 1 x 1 x 1, acceleration factor 2 in slice and phase directions, acquisition time = 5:03 min; and 3) 3D FLAIR: TR = 4800, TE = 271–415, TI = 1600, voxel size = 0.9 x 0.9 x 1, acceleration factor 2 in slice direction, 2.6 in phase direction, acquisition time = 6:10 min. All scans were acquired sagittally.

Image Preprocessing

For each individual subject, T2 and FLAIR images were co-registered to the T1 with an affine transformation using a normalized mutual information cost function and resampled to the T1 grid using AFNI.  Registered images were visually assessed for alignment.  Cortical reconstruction was performed using T1 and T2 images as input to FreeSurfer's standard processing pipeline with  FreeSurfer v6.0.0  \cite{Dale1999,Fischl1999}.  Results were visually inspected and manually corrected as needed. Cortical surfaces were resampled to a standard mesh using the AFNI SUMA package \cite{Cox1996a} to allow comparison of corresponding vertices across individuals.  Intensity correction was implemented using an in-house edge-aware intensity correction procedure (see supplementary methods for details, Figure S1).