Neuroimaging phenotyping and structural-metabolic-epileptogenic correlation of temporal neocortex in focal cortical dysplasia ?a
Abstract number :
417
Submission category :
5. Neuro Imaging / 5A. Structural Imaging
Year :
2020
Submission ID :
2422761
Source :
www.aesnet.org
Presentation date :
12/6/2020 12:00:00 PM
Published date :
Nov 21, 2020, 02:24 AM
Authors :
Jiajie Mo, Beijing Tiantan Hospital; Sophie Adler - University College London; Kai Zhang - Beijing Tiantan Hospital;;
Rationale:
Focal cortical dysplasia Ⅲa (FCD Ⅲa) is a common pathological subtype in temporal lobe epilepsy (TLE) and is difficult to diagnosis and surgical resection. The present study aimed to explore neuroimaging phenotyping and structural-metabolic-epileptogenic correlation of temporal neocortex in FCD Ⅲa.
Method:
In 64 FCD Ⅲa patients and 89 healthy controls, we performed the surface-based analysis to illustrate the neuroimaging phenotyping. LASSO-Logistic regression algorithm was designed to classify the FCD Ⅲa. Exploration of structural-metabolic-epileptogenic correlation evaluated the utility of neuroimaging alterations to predict epileptogenicity and postsurgical seizure outcomes.
Results:
Relative to controls, morphological changes of FCD Ⅲa were subtle but it displayed widespread hyperintensity in superior temporal lobe and mesial temporal neocortex that was most marked in parahippocampal gyrus. And the degree of abnormalities increased during the duration of disease. The algorithm successfully guided automated FCD Ⅲa classification (Control: 92.1%; FCD Ⅲa: 84.4%; FCD Ⅰ: 92.9%). The cross-modalities analysis shown a significantly negative correlation between FLAIR hyperintensity and PET hypometabolism. And the epileptogenic cortices were mostly located in the brain regions with simultaneously structural and metabolic abnormalities.
Conclusion:
Neuroimaging changes in temporal neocortex of FCD Ⅲa were subtle. While the computational analysis could help to identify the abnormalities and improve the diagnosis. Non-invasive structural alterations had the potential to predict the metabolic and epileptogenic state, which was conducive to reduce the radiative and invasive methods. But how to guide the surgery was inconclusive.
Funding:
:National Natural Science Foundation of China (No. 81771399, 81701276, 81641053).
Neuro Imaging