Negar Memarian1, Sally Kim2, Sandra Dewar3, Jerome Engel4, Richard J Staba2. 1. Department of Psychology, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States; Department of Neurology, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States. Electronic address: nmemarian@ucla.edu. 2. Department of Neurology, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States. 3. Department of Neurosurgery, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States. 4. Department of Neurology, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States; Department of Neurosurgery, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States; Department of Neurobiology, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States; Department of Psychiatry & Biobehavioral Sciences, David Geffen School of Medicine and at UCLA, Los Angeles, CA 90095, United States.
Abstract
BACKGROUND: This study sought to predict postsurgical seizure freedom from pre-operative diagnostic test results and clinical information using a rapid automated approach, based on supervised learning methods in patients with drug-resistant focal seizures suspected to begin in temporal lobe. METHOD: We applied machine learning, specifically a combination of mutual information-based feature selection and supervised learning classifiers on multimodal data, to predict surgery outcome retrospectively in 20 presurgical patients (13 female; mean age±SD, in years 33±9.7 for females, and 35.3±9.4 for males) who were diagnosed with mesial temporal lobe epilepsy (MTLE) and subsequently underwent standard anteromesial temporal lobectomy. The main advantage of the present work over previous studies is the inclusion of the extent of ipsilateral neocortical gray matter atrophy and spatiotemporal properties of depth electrode-recorded seizures as training features for individual patient surgery planning. RESULTS: A maximum relevance minimum redundancy (mRMR) feature selector identified the following features as the most informative predictors of postsurgical seizure freedom in this study's sample of patients: family history of epilepsy, ictal EEG onset pattern (positive correlation with seizure freedom), MRI-based gray matter thickness reduction in the hemisphere ipsilateral to seizure onset, proportion of seizures that first appeared in ipsilateral amygdala to total seizures, age, epilepsy duration, delay in the spread of ipsilateral ictal discharges from site of onset, gender, and number of electrode contacts at seizure onset (negative correlation with seizure freedom). Using these features in combination with a least square support vector machine (LS-SVM) classifier compared to other commonly used classifiers resulted in very high surgical outcome prediction accuracy (95%). CONCLUSIONS: Supervised machine learning using multimodal compared to unimodal data accurately predicted postsurgical outcome in patients with atypical MTLE. Published by Elsevier Ltd.
BACKGROUND: This study sought to predict postsurgical seizure freedom from pre-operative diagnostic test results and clinical information using a rapid automated approach, based on supervised learning methods in patients with drug-resistant focal seizures suspected to begin in temporal lobe. METHOD: We applied machine learning, specifically a combination of mutual information-based feature selection and supervised learning classifiers on multimodal data, to predict surgery outcome retrospectively in 20 presurgical patients (13 female; mean age±SD, in years 33±9.7 for females, and 35.3±9.4 for males) who were diagnosed with mesial temporal lobe epilepsy (MTLE) and subsequently underwent standard anteromesial temporal lobectomy. The main advantage of the present work over previous studies is the inclusion of the extent of ipsilateral neocortical gray matter atrophy and spatiotemporal properties of depth electrode-recorded seizures as training features for individual patient surgery planning. RESULTS: A maximum relevance minimum redundancy (mRMR) feature selector identified the following features as the most informative predictors of postsurgical seizure freedom in this study's sample of patients: family history of epilepsy, ictal EEG onset pattern (positive correlation with seizure freedom), MRI-based gray matter thickness reduction in the hemisphere ipsilateral to seizure onset, proportion of seizures that first appeared in ipsilateral amygdala to total seizures, age, epilepsy duration, delay in the spread of ipsilateral ictal discharges from site of onset, gender, and number of electrode contacts at seizure onset (negative correlation with seizure freedom). Using these features in combination with a least square support vector machine (LS-SVM) classifier compared to other commonly used classifiers resulted in very high surgical outcome prediction accuracy (95%). CONCLUSIONS: Supervised machine learning using multimodal compared to unimodal data accurately predicted postsurgical outcome in patients with atypical MTLE. Published by Elsevier Ltd.
Authors: Jennifer A Ogren; Anatol Bragin; Charles L Wilson; Gil D Hoftman; Jack J Lin; Rebecca A Dutton; Tony A Fields; Arthur W Toga; Paul M Thompson; Jerome Engel; Richard J Staba Journal: Epilepsia Date: 2008-11-19 Impact factor: 5.864
Authors: S S Keller; U C Wieshmann; C E Mackay; C E Denby; J Webb; N Roberts Journal: J Neurol Neurosurg Psychiatry Date: 2002-12 Impact factor: 10.154
Authors: Eliane Kobayashi; Maria Daniela D'Agostino; Iscia Lopes-Cendes; Eva Andermann; François Dubeau; Carlos A M Guerreiro; André A Schenka; Luciano S Queiroz; André Olivier; Fernando Cendes; Frederick Andermann Journal: Epilepsia Date: 2003-08 Impact factor: 5.864
Authors: Matthew D Smyth; David D Limbrick; Jeffrey G Ojemann; John Zempel; Shenandoah Robinson; Donncha F O'Brien; Russell P Saneto; Monisha Goyal; Richard E Appleton; Francesco T Mangano; Tae Sung Park Journal: J Neurosurg Date: 2007-03 Impact factor: 5.115
Authors: Wesley T Kerr; Stefan T Nguyen; Andrew Y Cho; Edward P Lau; Daniel H Silverman; Pamela K Douglas; Navya M Reddy; Ariana Anderson; Jennifer Bramen; Noriko Salamon; John M Stern; Mark S Cohen Journal: Front Neurol Date: 2013-04-03 Impact factor: 4.003
Authors: Gyujoon Hwang; Veena A Nair; Jed Mathis; Cole J Cook; Rosaleena Mohanty; Gengyan Zhao; Neelima Tellapragada; Candida Ustine; Onyekachi O Nwoke; Charlene Rivera-Bonet; Megan Rozman; Linda Allen; Courtney Forseth; Dace N Almane; Peter Kraegel; Andrew Nencka; Elizabeth Felton; Aaron F Struck; Rasmus Birn; Rama Maganti; Lisa L Conant; Colin J Humphries; Bruce Hermann; Manoj Raghavan; Edgar A DeYoe; Jeffrey R Binder; Elizabeth Meyerand; Vivek Prabhakaran Journal: Brain Connect Date: 2019-03
Authors: John P Andrews; Simon Ammanuel; Jonathan Kleen; Ankit N Khambhati; Robert Knowlton; Edward F Chang Journal: Epilepsia Date: 2020-09-17 Impact factor: 5.864
Authors: Ida A Nissen; Cornelis J Stam; Elisabeth C W van Straaten; Viktor Wottschel; Jaap C Reijneveld; Johannes C Baayen; Philip C de Witt Hamer; Sander Idema; Demetrios N Velis; Arjan Hillebrand Journal: Front Neurol Date: 2018-08-07 Impact factor: 4.003
Authors: Kevin Bretonnel Cohen; Benjamin Glass; Hansel M Greiner; Katherine Holland-Bouley; Shannon Standridge; Ravindra Arya; Robert Faist; Diego Morita; Francesco Mangano; Brian Connolly; Tracy Glauser; John Pestian Journal: Biomed Inform Insights Date: 2016-05-22
Authors: Negar Memarian; Jared B Torre; Kate E Haltom; Annette L Stanton; Matthew D Lieberman Journal: Soc Cogn Affect Neurosci Date: 2017-09-01 Impact factor: 3.436
Authors: Peter N Taylor; Nishant Sinha; Yujiang Wang; Sjoerd B Vos; Jane de Tisi; Anna Miserocchi; Andrew W McEvoy; Gavin P Winston; John S Duncan Journal: Neuroimage Clin Date: 2018-01-31 Impact factor: 4.881