Quantitative surface analysis of combined MRI and PET enhances detection of focal cortical dysplasias

被引:45
作者
Tan, Yee-Leng [1 ,6 ]
Kim, Hosung [2 ]
Lee, Seunghyun [3 ]
Tihan, Tarik [1 ]
Ver Hoef, Lawrence [5 ]
Mueller, Susanne G. [4 ]
Barkovich, Anthony James [4 ]
Xu, Duan [4 ]
Knowlton, Robert [1 ]
机构
[1] Univ Calif San Francisco, Dept Neurol, San Francisco, CA USA
[2] Univ Southern Calif, Keck Sch Med USC, Lab Neuro Imaging, Los Angeles, CA USA
[3] Univ Calif San Francisco, Dept Radiol & Biomed Imaging, San Francisco, CA 94143 USA
[4] Seoul Natl Univ Hosp, Dept Radiol, Seoul, South Korea
[5] Univ Alabama Birmingham, Dept Neurol, Birmingham, AL USA
[6] Natl Neurosci Inst, Dept Neurol, Singapore, Singapore
关键词
Focal cortical dysplasia; FCD detection; MRI; FDG-PET; Surface-based feature modeling; Patch analysis; TEMPORAL-LOBE EPILEPSY; IMPROVES DETECTION; FDG-PET; PROGNOSTIC-FACTORS; SEGMENTATION; CLASSIFICATION; SURGERY; MATTER; IMAGES; CORTEX;
D O I
10.1016/j.neuroimage.2017.10.065
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Objective: Focal cortical dysplasias (FCDs) often cause pharmacoresistant epilepsy, and surgical resection can lead to seizure-freedom. Magnetic resonance imaging (MRI) and positron emission tomography (PET) play complementary roles in FCD identification/localization; nevertheless, many FCDs are small or subtle, and difficult to find on routine radiological inspection. We aimed to automatically detect subtle or visually-unidentifiable FCDs by building a classifier based on an optimized cortical surface sampling of combined MRI and PET features. Methods: Cortical surfaces of 28 patients with histopathologically-proven FCDs were extracted. Morphology and intensity-based features characterizing FCD lesions were calculated vertex-wise on each cortical surface, and fed to a 2-step (Support Vector Machine and patch-based) classifier. Classifier performance was assessed compared to manual lesion labels. Results: Our classifier using combined feature selections from MRI and PET outperformed both quantitative MRI and multimodal visual analysis in FCD detection (93% vs 82% vs 68%). No false positives were identified in the controls, whereas 3.4% of the vertices outside FCD lesions were also classified to be lesional ("extralesional clusters"). Patients with type I or IIa FCDs displayed a higher prevalence of extralesional clusters at an intermediate distance to the FCD lesions compared to type IIb FCDs (p < 0.05). The former had a correspondingly lower chance of positive surgical outcome (71% vs 91%). Conclusions: Machine learning with multimodal feature sampling can improve FCD detection. The spread of extralesional clusters characterize different FCD subtypes, and may represent structurally or functionally abnormal tissue on a microscopic scale, with implications for surgical outcomes.
引用
收藏
页码:10 / 18
页数:9
相关论文
共 41 条
[1]   Effects of hardware heterogeneity on the performance of SVM Alzheimer's disease classifier [J].
Abdulkadir, Ahmed ;
Mortamet, Benedicte ;
Vemuri, Prashanthi ;
Jack, Clifford R., Jr. ;
Krueger, Gunnar ;
Kloeppel, Stefan .
NEUROIMAGE, 2011, 58 (03) :785-792
[2]   Cortical feature analysis and machine learning improves detection of "MRI-negative" focal cortical dysplasia [J].
Ahmed, Bilal ;
Brodley, Carla E. ;
Blackmon, Karen E. ;
Kuzniecky, Ruben ;
Barash, Gilad ;
Carlson, Chad ;
Quinn, Brian T. ;
Doyle, Werner ;
French, Jacqueline ;
Devinsky, Orrin ;
Thesen, Thomas .
EPILEPSY & BEHAVIOR, 2015, 48 :21-28
[3]   Computational models of MRI characteristics of focal cortical dysplasia improve lesion detection [J].
Antel, SB ;
Bernasconi, A ;
Bernasconi, N ;
Collins, DL ;
Kearney, RE ;
Shinghal, R ;
Arnold, DL .
NEUROIMAGE, 2002, 17 (04) :1755-1760
[4]  
Besson P, 2008, LECT NOTES COMPUT SC, V5241, P645, DOI 10.1007/978-3-540-85988-8_77
[5]   Small focal cortical dysplasia lesions are located at the bottom of a deep sulcus [J].
Besson, Pierre ;
Andermann, Frederick ;
Dubeau, Francois ;
Bernasconi, Andrea .
BRAIN, 2008, 131 :3246-3255
[6]   Cortical Morphology in 6-to 10-Year Old Children With Autistic Traits: A Population-Based Neuroimaging Study [J].
Blanken, Laura M. E. ;
Mous, Sabine E. ;
Ghassabian, Akhgar ;
Muetzel, Ryan L. ;
Schoemaker, Nikita K. ;
El Marroun, Hanan ;
van der Lugt, Aad ;
Jaddoe, Vincent W. V. ;
Hofman, Albert ;
Verhulst, Frank C. ;
Tiemeier, Henning ;
White, Tonya .
AMERICAN JOURNAL OF PSYCHIATRY, 2015, 172 (05) :479-486
[7]   The clinicopathologic spectrum of focal cortical dysplasias: A consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission [J].
Bluemcke, Ingmar ;
Thom, Maria ;
Aronica, Eleonora ;
Armstrong, Dawna D. ;
Vinters, Harry V. ;
Palmini, Andre ;
Jacques, Thomas S. ;
Avanzini, Giuliano ;
Barkovich, A. James ;
Battaglia, Giorgio ;
Becker, Albert ;
Cepeda, Carlos ;
Cendes, Fernando ;
Colombo, Nadia ;
Crino, Peter ;
Cross, J. Helen ;
Delalande, Olivier ;
Dubeau, Francois ;
Duncan, John ;
Guerrini, Renzo ;
Kahane, Philippe ;
Mathern, Gary ;
Najm, Imad ;
Ozkara, Cigdem ;
Raybaud, Charles ;
Represa, Alfonso ;
Roper, Steven N. ;
Salamon, Noriko ;
Schulze-Bonhage, Andreas ;
Tassi, Laura ;
Vezzani, Annamaria ;
Spreafico, Roberto .
EPILEPSIA, 2011, 52 (01) :158-174
[8]   Depth potential function for folding pattern representation, registration and analysis [J].
Boucher, Maxime ;
Whitesides, Sue ;
Evans, Alan .
MEDICAL IMAGE ANALYSIS, 2009, 13 (02) :203-214
[9]   FDG-PET improves surgical outcome in negative MRI Taylor-type focal cortical dysplasias [J].
Chassoux, F. ;
Rodrigo, S. ;
Semah, F. ;
Beuvon, F. ;
Landre, E. ;
Devaux, B. ;
Turak, B. ;
Mellerio, C. ;
Meder, J. -F. ;
Roux, F. -X. ;
Daumas-Duport, C. ;
Merlet, P. ;
Dulac, O. ;
Chiron, C. .
NEUROLOGY, 2010, 75 (24) :2168-2175
[10]   AUTOMATIC 3D INTERSUBJECT REGISTRATION OF MR VOLUMETRIC DATA IN STANDARDIZED TALAIRACH SPACE [J].
COLLINS, DL ;
NEELIN, P ;
PETERS, TM ;
EVANS, AC .
JOURNAL OF COMPUTER ASSISTED TOMOGRAPHY, 1994, 18 (02) :192-205