Localization of epileptic focus by gray matter reduction analysis from brain MR images for temporal lobe epilepsy patients

被引:1
作者
Ficici, Cansel [1 ]
Telatar, Ziya [2 ]
Erogul, Osman [3 ]
机构
[1] Ankara Univ, Dept Elect & Elect Engn, TR-06830 Ankara, Turkiye
[2] Baskent Univ, Dept Biomed Engn, TR-06790 Ankara, Turkiye
[3] TOBB Univ Econ & Technol, Dept Biomed Engn, TR-06560 Ankara, Turkiye
关键词
MRI; Temporal lobe epilepsy; Voxel based morphometry; Epileptic focus; SCLEROSIS; CLASSIFICATION;
D O I
10.1016/j.bspc.2023.104716
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
Localization of epileptic focus is crucial for resective epilepsy surgery and treatment planning. The purpose of this study is to develop a method analyzing gray matter reduction in brain magnetic resonance images in order to identify epileptogenic focus of temporal lobe epilepsy (TLE) patients. So, a new voxel based morphometry analysis based epileptogenic brain side detection approach was proposed. Gray matter abnormalities were detected from T1-weighted MR images by using Statistical Parametric Mapping based voxel based morphometry analysis. The dataset of the introduced retrospective analysis consists of MR images of 15 TLE patients including patients with hippocampal sclerosis, mesial temporal sclerosis, and MRI negative diagnoses. In addition, MRI of 14 healthy subjects were used as the control group. TLE focus detection performed by the proposed method and seizure lateralization from EEG recordings realized by the expert overlapped at a rate of 91.7 %. In addition, sensitivity of 100 % and 80 % were obtained for right TLE and left TLE detection, respectively. Experimental results showed that the proposed algorithm can reveal subtle Gray matter reduction in the temporal lobe and limbic lobe areas, thus providing an automated medical support system for the expert in identifying the epileptic focus of TLE patients.
引用
收藏
页数:8
相关论文
共 18 条
[1]  
alivelearn.net, XJVIEW TOOLB
[2]  
Alvarez-Linera J., 2019, Clinical Neuroradiology: The ESNR Textbook, P891, DOI DOI 10.1007/978-3-319-68536-6_50
[3]   Unified segmentation [J].
Ashburner, J ;
Friston, KJ .
NEUROIMAGE, 2005, 26 (03) :839-851
[4]   A fast diffeomorphic image registration algorithm [J].
Ashburner, John .
NEUROIMAGE, 2007, 38 (01) :95-113
[5]   Accurate lateralization and classification of MRI-negative 18F-FDG-PET--positive temporal lobe epilepsy using double inversion recovery and machine-learning [J].
Beheshti, Iman ;
Sone, Daichi ;
Maikusa, Norihide ;
Kimura, Yukio ;
Shigemoto, Yoko ;
Sato, Noriko ;
Matsuda, Hiroshi .
COMPUTERS IN BIOLOGY AND MEDICINE, 2021, 137
[6]   Gray Matter and White Matter Abnormalities in Temporal Lobe Epilepsy Patients with and without Hippocampal Sclerosis [J].
Beheshti, Iman ;
Sone, Daichi ;
Farokhian, Farnaz ;
Maikusa, Norihide ;
Matsuda, Hiroshi .
FRONTIERS IN NEUROLOGY, 2018, 9
[7]   International consensus classification of hippocampal sclerosis in temporal lobe epilepsy: A Task Force report from the ILAE Commission on Diagnostic Methods [J].
Bluemcke, Ingmar ;
Thom, Maria ;
Aronica, Eleonora ;
Armstrong, Dawna D. ;
Bartolomei, Fabrice ;
Bernasconi, Andrea ;
Bernasconi, Neda ;
Bien, Christian G. ;
Cendes, Fernando ;
Coras, Roland ;
Cross, J. Helen ;
Jacques, Thomas S. ;
Kahane, Philippe ;
Mathern, Gary W. ;
Miyata, Haijme ;
Moshe, Solomon L. ;
Oz, Buge ;
Oezkara, Cigdem ;
Perucca, Emilio ;
Sisodiya, Sanjay ;
Wiebe, Samuel ;
Spreafico, Roberto .
EPILEPSIA, 2013, 54 (07) :1315-1329
[8]   Limbic system abnormalities associated with mesial temporal sclerosis: A model of chronic cerebral changes due to seizures [J].
Chan, S ;
Erickson, JK ;
Yoon, SS .
RADIOGRAPHICS, 1997, 17 (05) :1095-1110
[9]   Voxel-based morphometry analysis and machine learning based classification in pediatric mesial temporal lobe epilepsy with hippocampal sclerosis [J].
Chen, Shihui ;
Zhang, Jian ;
Ruan, Xiaolei ;
Deng, Kan ;
Zhang, Jianing ;
Zou, Dongfang ;
He, Xiaoming ;
Li, Feng ;
Bin, Guo ;
Zeng, Hongwu ;
Huang, Bingsheng .
BRAIN IMAGING AND BEHAVIOR, 2020, 14 (05) :1945-1954
[10]  
fil.ion.ucl, SPM SOFTW