Yee-Leng Tan1, Hosung Kim2, Seunghyun Lee3, Tarik Tihan4, Lawrence Ver Hoef5, Susanne G Mueller6, Anthony James Barkovich7, Duan Xu8, Robert Knowlton9. 1. Department of Neurology, University of California, San Francisco, San Francisco, CA, USA; Department of Neurology, National Neuroscience Institute, Singapore. Electronic address: tan.yee.leng.neuro@singhealth.com.sg. 2. Laboratory of Neuro Imaging, Keck School of Medicine of USC, University of Southern California, Los Angeles, CA, USA. Electronic address: hosung.kim@loni.usc.edu. 3. Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, USA. Electronic address: seunghyun.lee.22@gmail.com. 4. Department of Neurology, University of California, San Francisco, San Francisco, CA, USA. Electronic address: tarik.tihan@ucsf.edu. 5. Department of Neurology, University of Alabama, Birmingham, United Kingdom. Electronic address: lverhoef@uab.edu. 6. Department of Radiology, Seoul National University Hospital, Republic of Korea. Electronic address: susanne.mueller@ucsf.edu. 7. Department of Radiology, Seoul National University Hospital, Republic of Korea. Electronic address: james.barkovich@ucsf.edu. 8. Department of Radiology, Seoul National University Hospital, Republic of Korea. Electronic address: duan.xu@ucsf.edu. 9. Department of Neurology, University of California, San Francisco, San Francisco, CA, USA. Electronic address: robert.knowlton@ucsf.edu.
Abstract
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.
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.
Authors: Hans-Jürgen Huppertz; Christina Grimm; Susanne Fauser; Jan Kassubek; Irina Mader; Albrecht Hochmuth; Joachim Spreer; Andreas Schulze-Bonhage Journal: Epilepsy Res Date: 2005-09-19 Impact factor: 3.045
Authors: June Sic Kim; Vivek Singh; Jun Ki Lee; Jason Lerch; Yasser Ad-Dab'bagh; David MacDonald; Jong Min Lee; Sun I Kim; Alan C Evans Journal: Neuroimage Date: 2005-08-01 Impact factor: 6.556
Authors: Laura M E Blanken; Sabine E Mous; Akhgar Ghassabian; Ryan L Muetzel; Nikita K Schoemaker; Hanan El Marroun; Aad van der Lugt; Vincent W V Jaddoe; Albert Hofman; Frank C Verhulst; Henning Tiemeier; Tonya White Journal: Am J Psychiatry Date: 2015-01-13 Impact factor: 18.112
Authors: Ingmar Blümcke; Maria Thom; Eleonora Aronica; Dawna D Armstrong; Harry V Vinters; Andre Palmini; Thomas S Jacques; Giuliano Avanzini; A James Barkovich; Giorgio Battaglia; Albert Becker; Carlos Cepeda; Fernando Cendes; Nadia Colombo; Peter Crino; J Helen Cross; Olivier Delalande; François Dubeau; John Duncan; Renzo Guerrini; Philippe Kahane; Gary Mathern; Imad Najm; Ciğdem Ozkara; Charles Raybaud; Alfonso Represa; Steven N Roper; Noriko Salamon; Andreas Schulze-Bonhage; Laura Tassi; Annamaria Vezzani; Roberto Spreafico Journal: Epilepsia Date: 2010-11-10 Impact factor: 5.864
Authors: Sang Kun Lee; Seo Young Lee; Kwang-Ki Kim; Kkeun-Sik Hong; Dong-Soo Lee; Chun-Kee Chung Journal: Ann Neurol Date: 2005-10 Impact factor: 10.422