NeuroMorphix: A Novel Brain MRI Asymmetry-specific Feature Construction Approach For Seizure Recurrence Prediction

TL;DR

Seizure recurrence prediction after a first unprovoked seizure is clinically critical but current risk predictors are imperfect, leading to unnecessary treatment or preventable seizures. The authors introduce NeuroMorphix, a framework that transforms FreeSurfer-derived morphometrics from clinical 3T MRI into whole-brain asymmetry features across cortical and subcortical regions and trains multiple classifiers. In a first-seizure cohort (n=169), top NeuroMorphix features yield AUROC values of 88–93%, accuracies of 83–89%, and F1 scores of 83–90%, with highly ranked features reflecting known epilepsy-related structural alterations. The approach provides an interpretable, data-driven method to support clinical decisions and has potential applicability to other brain disorders characterized by hemispheric asymmetry.

Abstract

Seizure recurrence is an important concern after an initial unprovoked seizure; without drug treatment, it occurs within 2 years in 40-50% of cases. The decision to treat currently relies on predictors of seizure recurrence risk that are inaccurate, resulting in unnecessary, possibly harmful, treatment in some patients and potentially preventable seizures in others. Because of the link between brain lesions and seizure recurrence, we developed a recurrence prediction tool using machine learning and clinical 3T brain MRI. We developed NeuroMorphix, a feature construction approach based on MRI brain anatomy. Each of seven NeuroMorphix features measures the absolute or relative difference between corresponding regions in each cerebral hemisphere. FreeSurfer was used to segment brain regions and to generate values for morphometric parameters (8 for each cortical region and 5 for each subcortical region). The parameters were then mapped to whole brain NeuroMorphix features, yielding a total of 91 features per subject. Features were generated for a first seizure patient cohort (n = 169) categorised into seizure recurrence and non-recurrence subgroups. State-of-the-art classification algorithms were trained and tested using NeuroMorphix features to predict seizure recurrence. Classification models using the top 5 features, ranked by sequential forward selection, demonstrated excellent performance in predicting seizure recurrence, with area under the ROC curve of 88-93%, accuracy of 83-89%, and F1 score of 83-90%. Highly ranked features aligned with structural alterations known to be associated with epilepsy. This study highlights the potential for targeted, data-driven approaches to aid clinical decision-making in brain disorders.

Figures (3)

• Figure 1: Proposed framework for seizure recurrence prediction using proposed NeuroMorphix features.
• Figure 2: Shown are feature selection results using the (a) sequential feature selection and (b) sequential backward elimination approaches. The horizontal axis represents the number of features used to achieve that level of accuracy (vertical axis). Colours correspond with the different classification algorithms evaluated.
• Figure 3: Feature ranking based on classification accuracy. The lower number and dark colour represent higher rank whereas higher number and brighter colour represent low rank, respectively.