Bo Jin1,2, Balu Krishnan2, Sophie Adler3,4, Konrad Wagstyl3,5, Wenhan Hu6, Stephen Jones7, Imad Najm2, Andreas Alexopoulos2, Kai Zhang6, Jianguo Zhang6, Meiping Ding1, Shuang Wang1, Zhong Irene Wang2. 1. Department of Neurology, School of Medicine, Epilepsy Center, Second Affiliated Hospital, Zhejiang University, Hangzhou, China. 2. Epilepsy Center, Cleveland Clinic, Cleveland, OH, USA. 3. Developmental Neurosciences, UCL Great Ormond Street Institute of Child Health, University College London, London, UK. 4. Great Ormond Street Hospital for Children, London, UK. 5. Brain Mapping Unit, Institute of Psychiatry, University of Cambridge, Cambridge, UK. 6. Department of Neurosurgery, Beijing Neurosurgical Institute, Beijing Tiantan Hospital, Capital Medical University, Beijing, China. 7. Imaging Institute, Cleveland Clinic, Cleveland, OH, USA.
Abstract
OBJECTIVE: Focal cortical dysplasia (FCD) is a major pathology in patients undergoing surgical resection to treat pharmacoresistant epilepsy. Magnetic resonance imaging (MRI) postprocessing methods may provide essential help for detection of FCD. In this study, we utilized surface-based MRI morphometry and machine learning for automated lesion detection in a mixed cohort of patients with FCD type II from 3 different epilepsy centers. METHODS: Sixty-one patients with pharmacoresistant epilepsy and histologically proven FCD type II were included in the study. The patients had been evaluated at 3 different epilepsy centers using 3 different MRI scanners. T1-volumetric sequence was used for postprocessing. A normal database was constructed with 120 healthy controls. We also included 35 healthy test controls and 15 disease test controls with histologically confirmed hippocampal sclerosis to assess specificity. Features were calculated and incorporated into a nonlinear neural network classifier, which was trained to identify lesional cluster. We optimized the threshold of the output probability map from the classifier by performing receiver operating characteristic (ROC) analyses. Success of detection was defined by overlap between the final cluster and the manual labeling. Performance was evaluated using k-fold cross-validation. RESULTS: The threshold of 0.9 showed optimal sensitivity of 73.7% and specificity of 90.0%. The area under the curve for the ROC analysis was 0.75, which suggests a discriminative classifier. Sensitivity and specificity were not significantly different for patients from different centers, suggesting robustness of performance. Correct detection rate was significantly lower in patients with initially normal MRI than patients with unequivocally positive MRI. Subgroup analysis showed the size of the training group and normal control database impacted classifier performance. SIGNIFICANCE: Automated surface-based MRI morphometry equipped with machine learning showed robust performance across cohorts from different centers and scanners. The proposed method may be a valuable tool to improve FCD detection in presurgical evaluation for patients with pharmacoresistant epilepsy. Wiley Periodicals, Inc.
OBJECTIVE:Focal cortical dysplasia (FCD) is a major pathology in patients undergoing surgical resection to treat pharmacoresistant epilepsy. Magnetic resonance imaging (MRI) postprocessing methods may provide essential help for detection of FCD. In this study, we utilized surface-based MRI morphometry and machine learning for automated lesion detection in a mixed cohort of patients with FCD type II from 3 different epilepsy centers. METHODS: Sixty-one patients with pharmacoresistant epilepsy and histologically proven FCD type II were included in the study. The patients had been evaluated at 3 different epilepsy centers using 3 different MRI scanners. T1-volumetric sequence was used for postprocessing. A normal database was constructed with 120 healthy controls. We also included 35 healthy test controls and 15 disease test controls with histologically confirmed hippocampal sclerosis to assess specificity. Features were calculated and incorporated into a nonlinear neural network classifier, which was trained to identify lesional cluster. We optimized the threshold of the output probability map from the classifier by performing receiver operating characteristic (ROC) analyses. Success of detection was defined by overlap between the final cluster and the manual labeling. Performance was evaluated using k-fold cross-validation. RESULTS: The threshold of 0.9 showed optimal sensitivity of 73.7% and specificity of 90.0%. The area under the curve for the ROC analysis was 0.75, which suggests a discriminative classifier. Sensitivity and specificity were not significantly different for patients from different centers, suggesting robustness of performance. Correct detection rate was significantly lower in patients with initially normal MRI than patients with unequivocally positive MRI. Subgroup analysis showed the size of the training group and normal control database impacted classifier performance. SIGNIFICANCE: Automated surface-based MRI morphometry equipped with machine learning showed robust performance across cohorts from different centers and scanners. The proposed method may be a valuable tool to improve FCD detection in presurgical evaluation for patients with pharmacoresistant epilepsy. Wiley Periodicals, Inc.
Authors: Bruce Fischl; David H Salat; Evelina Busa; Marilyn Albert; Megan Dieterich; Christian Haselgrove; Andre van der Kouwe; Ron Killiany; David Kennedy; Shuna Klaveness; Albert Montillo; Nikos Makris; Bruce Rosen; Anders M Dale Journal: Neuron Date: 2002-01-31 Impact factor: 17.173
Authors: A Bernasconi; S B Antel; D L Collins; N Bernasconi; A Olivier; F Dubeau; G B Pike; F Andermann; D L Arnold Journal: Ann Neurol Date: 2001-06 Impact factor: 10.422
Authors: Bilal Ahmed; Carla E Brodley; Karen E Blackmon; Ruben Kuzniecky; Gilad Barash; Chad Carlson; Brian T Quinn; Werner Doyle; Jacqueline French; Orrin Devinsky; Thomas Thesen Journal: Epilepsy Behav Date: 2015-05-31 Impact factor: 2.937
Authors: Francesco Cardinale; Stefano Francione; Luciana Gennari; Alberto Citterio; Maurizio Sberna; Laura Tassi; Roberto Mai; Ivana Sartori; Lino Nobili; Massimo Cossu; Laura Castana; Giorgio Lo Russo; Nadia Colombo Journal: World Neurosurg Date: 2016-11-25 Impact factor: 2.104
Authors: Wendy A Chamberlain; Mark L Cohen; Kymberly A Gyure; Bette K Kleinschmidt-DeMasters; Arie Perry; Suzanne Z Powell; Jiang Qian; Susan M Staugaitis; Richard A Prayson Journal: Epilepsia Date: 2009-10-08 Impact factor: 5.864
Authors: Thomas Thesen; Brian T Quinn; Chad Carlson; Orrin Devinsky; Jonathan DuBois; Carrie R McDonald; Jacqueline French; Richard Leventer; Olga Felsovalyi; Xiuyuan Wang; Eric Halgren; Ruben Kuzniecky Journal: PLoS One Date: 2011-02-04 Impact factor: 3.240
Authors: Sophie Adler; Sara Lorio; Thomas S Jacques; Barbora Benova; Roxana Gunny; J Helen Cross; Torsten Baldeweg; David W Carmichael Journal: Neuroimage Clin Date: 2017-04-20 Impact factor: 4.881
Authors: Irene Wang; Sehong Oh; Ingmar Blümcke; Roland Coras; Balu Krishnan; Sanghoon Kim; Aaron McBride; Olesya Grinenko; Yicong Lin; Margit Overmyer; Tin Tun Aung; Mark Lowe; Mykol Larvie; Andreas V Alexopoulos; William Bingaman; Jorge A Gonzalez-Martinez; Imad Najm; Stephen E Jones Journal: Epilepsia Date: 2020-09-19 Impact factor: 5.864
Authors: Patrick H Luckett; Luigi Maccotta; John J Lee; Ki Yun Park; Nico U F Dosenbach; Beau M Ances; Robert Edward Hogan; Joshua S Shimony; Eric C Leuthardt Journal: Epilepsia Date: 2022-04-01 Impact factor: 6.740
Authors: Shan Wang; Bo Jin; Thandar Aung; Masaya Katagiri; Stephen E Jones; Balu Krishnan; Jorge A Gonzalez-Martinez; Richard A Prayson; Imad M Najm; Andreas V Alexopoulos; Shuang Wang; Meiping Ding; Zhong Irene Wang Journal: Front Neurol Date: 2018-11-27 Impact factor: 4.003