A Machine Learning Model to Predict Seizure Susceptibility from Resting-State fMRI Connectivity

Traumatic brain injury (TBI) is a leading cause of disability globally. Many patients develop post-traumatic epilepsy, or recurrent seizures following TBI. In recent years, significant efforts have been made to identify biomarkers of epileptogenesis that may assist in preventing seizure occurrence by identifying high-risk patients. We present a novel method of assessing seizure susceptibility using data from 49 patients enrolled in the Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx). We employ a machine learning paradigm that utilizes a Random Forest classifier trained with resting-state functional magnetic resonance imaging (fMRI) data to predict seizure outcomes. Following 100 rounds of stratified cross-validation with 70% of resting state fMRI scans as the training set and 30% as the testing set, our model was found to assess seizure outcome in the testing set with 69% accuracy. To validate the method, we compared our results with classification by Support Vector Machines and Neural Network classifiers.

[1]  Flavia Carle,et al.  Predicting Posttraumatic Epilepsy with MRI: Prospective Longitudinal Morphologic Study in Adults , 2005, Epilepsia.

[2]  Waqas Majeed,et al.  Functional Connectivity Alterations in Epilepsy from Resting-State Functional MRI , 2015, PloS one.

[3]  D C Good,et al.  Changes in resting connectivity during recovery from severe traumatic brain injury. , 2011, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[4]  Asla Pitkänen,et al.  MRI biomarkers for post-traumatic epileptogenesis. , 2013, Journal of neurotrauma.

[5]  Andreas Schulze-Bonhage,et al.  Epileptic seizure predictors based on computational intelligence techniques: A comparative study with 278 patients , 2014, Comput. Methods Programs Biomed..

[6]  Hal Blumenfeld,et al.  Neuroimaging biomarkers of epileptogenesis , 2011, Neuroscience Letters.

[7]  Yu. V. Obukhov,et al.  Epileptiform Activity Detection and Classification Algorithms of Rats with Post-traumatic Epilepsy , 2018 .

[8]  Terence O'Brien,et al.  Epileptic Seizure Prediction Using Big Data and Deep Learning: Toward a Mobile System , 2017, EBioMedicine.

[9]  Yufeng Zang,et al.  DPARSF: A MATLAB Toolbox for “Pipeline” Data Analysis of Resting-State fMRI , 2010 .

[10]  C. Teixeira,et al.  Preprocessing effects of 22 linear univariate features on the performance of seizure prediction methods , 2013, Journal of Neuroscience Methods.

[11]  Adel K. Afifi,et al.  Functional Neuroanatomy: Text and Atlas , 1997 .

[12]  Nitesh V. Chawla,et al.  Automated epileptic seizure detection using improved correlation-based feature selection with random forest classifier , 2017, Neurocomputing.

[13]  Gregory A. Worrell,et al.  Electrophysiological Biomarkers of Epilepsy , 2014, Neurotherapeutics.

[14]  Wei Cheng,et al.  Pattern Classification of Large-Scale Functional Brain Networks: Identification of Informative Neuroimaging Markers for Epilepsy , 2012, PloS one.

[15]  Brian Litt,et al.  The effect of increased intracranial EEG sampling rates in clinical practice , 2018, Clinical Neurophysiology.

[16]  Sarah M. Rajtmajer,et al.  The Rich Get Richer: Brain Injury Elicits Hyperconnectivity in Core Subnetworks , 2014, PloS one.

[17]  Bekir Karlik,et al.  Comparison Machine Learning Algorithms for Recognition of Epileptic Seizures in EEG , 2014, IWBBIO.

[18]  Karl J. Friston,et al.  Statistical parametric maps in functional imaging: A general linear approach , 1994 .

[19]  Shengwen Guo,et al.  A Comparative Study of Feature Selection Methods for the Discriminative Analysis of Temporal Lobe Epilepsy , 2017, Front. Neurol..

[20]  Zhifeng Kou,et al.  Compensation through Functional Hyperconnectivity: A Longitudinal Connectome Assessment of Mild Traumatic Brain Injury , 2015, Neural plasticity.

[21]  Evangelia Pippa,et al.  EEG-based Classification of Epileptic and Non-Epileptic Events using Multi-Array Decomposition , 2016, Int. J. Monit. Surveillance Technol. Res..

[22]  Lauren C. Frey,et al.  Epidemiology of Posttraumatic Epilepsy: A Critical Review , 2003, Epilepsia.

[23]  Fahmeed Hyder,et al.  Decreased Resting Functional Connectivity after Traumatic Brain Injury in the Rat , 2014, PloS one.

[24]  Steven Macey,et al.  The costs of traumatic brain injury: a literature review , 2013, ClinicoEconomics and outcomes research : CEOR.

[25]  Dhiya Al-Jumeily,et al.  A machine learning system for automated whole-brain seizure detection , 2016 .

[26]  Dinggang Shen,et al.  Evaluation of machine learning algorithms for treatment outcome prediction in patients with epilepsy based on structural connectome data , 2015, NeuroImage.

[27]  Simon Fong,et al.  Epileptic Seizures Prediction Using Machine Learning Methods , 2017, Comput. Math. Methods Medicine.

[28]  Yong He,et al.  BrainNet Viewer: A Network Visualization Tool for Human Brain Connectomics , 2013, PloS one.

[29]  Mojtaba Bandarabadi,et al.  Epileptic seizure prediction using relative spectral power features , 2015, Clinical Neurophysiology.

[30]  Nicola Amoroso,et al.  Multiplex Networks for Early Diagnosis of Alzheimer's Disease , 2018, Front. Aging Neurosci..

[31]  Fabio Ribeiro Cerqueira,et al.  Patient-Specific Epilepsy Seizure Detection Using Random Forest Classification over One-Dimension Transformed EEG Data , 2016, ISDA.