Ümit Özgür AKDEMİR, Seda GÜLBAHAR, Kerim ŞEKER, Uğuray AYDOS, Neşe İlgin KARABACAK, Lütfiye Özlem ATAY, Gökhan KURT, İrem YILDIRIM, Erhan BİLİR

Evaluation of brain FDG PET images in temporal lobe epilepsy for lateralization of epileptogenic focus using data mining methods

Background/aim: In temporal lobe epilepsy (TLE), brain positron emission tomography (PET) performed with F-18 fluorodeoxyglucose(FDG) is commonly used for lateralization of the epileptogenic temporal lobe. In this study, we aimed to evaluate the success ofquantitative analysis of brain FDG PET images using data mining methods in the lateralization of the epileptogenic temporal lobe.Materials and methods: Presurgical interictal brain FDG PET images of 49 adult mesial TLE patients with a minimum of 2 years ofpostsurgical follow-up and Engel I outcomes were retrospectively analyzed. Asymmetry indices were calculated from PET images fromthe mesial temporal lobe and its contiguous structures. The J48 and the logistic model tree (LMT) data mining algorithms were used tofind classification rules for the lateralization of the epileptogenic temporal lobe. The classification results obtained by these rules werecompared with the physicians’ visual readings and the findings of single-patient statistical parametric mapping (SPM) analyses in a testset of 18 patients. An additional 5-fold cross-validation was applied to the data to overcome the limitation of a relatively small samplesize.Results: In the lateralization of 18 patients in the test set, J48 and LMT methods were successful in 16 (89%) and 17 (94%) patients,respectively. The visual consensus readings were correct in all patients and SPM results were correct in 16 patients. The 5-fold crossvalidation method resulted in a mean correct lateralization ratio of 96% (47/49) for the LMT algorithm. This ratio was 88% (43 / 49) forthe J48 algorithm.Conclusion: Lateralization of the epileptogenic temporal lobe with data mining methods using regional metabolic asymmetry valuesobtained from interictal brain FDG PET images in mesial TLE patients is highly accurate. The application of data mining can contributeto the reader in the process of visual evaluation of FDG PET images of the brain.

PDF

___

1. Theodore WH. Presurgical focus localization in epilepsy: PET and SPECT. Seminars in Nuclear Medicine 2017; 47 (1): 44-53. doi: 10.1053/j.semnuclmed.2016.09.008
2. Kumar A, Chugani HT. The role of radionuclide imaging in epilepsy, Part 1: Sporadic temporal and extratemporal lobe epilepsy. Journal of Nuclear Medicine Technology 2017; 45 (1): 14-21. doi: 10.2967/jnumed.112.114397
3. Van Bogaert P, Massager N, Tugendhaft P, Wikler D, Damhaut P et al. Statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy. Neuroimage 2000; 12 (2): 129-138. doi: 10.1006/nimg.2000.0606
4. Willmann O, Wennberg R, May T, Woermann FG, PohlmannEden B. The contribution of 18F-FDG PET in preoperative epilepsy surgery evaluation for patients with temporal lobe epilepsy: A meta-analysis. Seizure-European Journal of Epilepsy 2007; 16 (6): 509-520. doi: 10.1016/j.seizure.2007.04.001
5. Soma T, Momose T, Takahashi M, Koyama K, Kawai K et al. Usefulness of extent analysis for statistical parametric mapping with asymmetry index using inter-ictal FGD-PET in mesial temporal lobe epilepsy. Annals of Nuclear Medicine 2012; 26 (4): 319-326. doi: 10.1007/s12149-012-0573-8
6. Takahashi M, Soma T, Kawai K, Koyama K, Ohtomo K et al. Voxel-based comparison of preoperative FDG-PET between mesial temporal lobe epilepsy patients with and without postoperative seizure-free outcomes. Annals of Nuclear Medicine 2012; 26 (9): 698-706. doi: 10.1007/s12149-012- 0629-9
7. Yang PF, Pei JS, Zhang HJ, Lin Q, Mei Z et al. Long-term epilepsy surgery outcomes in patients with PET-positive, MRInegative temporal lobe epilepsy. Epilepsy & Behavior 2014; 41: 91-97. doi: 10.1016/j.yebeh.2014.09.054
8. Capraz IY, Kurt G, Akdemir Ö, Hirfanoglu T, Oner Y et al. Surgical outcome in patients with MRI-negative, PET-positive temporal lobe epilepsy. Seizure-European Journal of Epilepsy 2015; 29: 63-68. doi: 10.1016/j.seizure.2015.03.015
9. Peter J, Houshmand S, Werner TJ, Rubello D, Alavi A. Novel assessment of global metabolism by F-18-FDG-PET for localizing affected lobe in temporal lobe epilepsy. Nuclear Medicine Communications 2016; 37 (8): 882-887. doi: 10.1097/ MNM.0000000000000526
10. Akman CI, Ichise M, Olsavsky A, Tikofsky RS, Van Heerturn RL et al. Epilepsy duration impacts on brain glucose metabolism in temporal lobe epilepsy: Results of voxel-based mapping. Epilepsy & Behavior 2010; 17 (3): 373-380. doi: 10.1016/j. yebeh.2009.12.007
11. Ohta Y, Nariai T, Ishii K, Ishiwata K, Mishina M et al. Annals Voxel- and ROI-based statistical analyses of PET parameters for guidance in the surgical treatment of intractable mesial temporal lobe epilepsy. Nuclear Medicine 2008; 22 (6): 495- 503. doi: 10.1007/s12149-008-0140-5
12. van’t Klooster MA, Huiskamp G, Zijlmans M, Debets RM, Comans EF et al. Can we increase the yield of FDG-PET in the preoperative work-up for epilepsy surgery? Epilepsy Research 2014; 108 (6): 1095-1105. doi: 10.1016/j.eplepsyres.2014.04.011
13. Rathore C, Dickson JC, Teotonio R, Ell P, Duncan JS. The utility of 18F-fluorodeoxyglucose PET (FDG PET) in epilepsy surgery. Epilepsy Research 2014; 108 (8): 1306-1314. doi: 10.1016/j.eplepsyres.2014.06.012
14. Kumar A, Juhasz C, Asano E, Sood S, Muzik O et al. Objective detection of epileptic foci by 18F-FDG PET in children undergoing epilepsy surgery. Journal of Nuclear Medicine 2010; 51 (12): 1901-1907. doi: 10.2967/jnumed.110.075390
15. Muzik O, Chugani DC, Shen C, da Silva EA, Shah J et al. Objective method for localization of cortical asymmetries using positron emission tomography to aid surgical resection of epileptic foci. Computer Aided Surgery 1998; 3 (2): 74-82. doi: 10.3109/10929089809148132
16. Chassoux F, Artiges E, Semah F, Desarnaud S, Laurent A et al. Determinants of brain metabolism changes in mesial temporal lobe epilepsy. Epilepsia 2016; 57 (6): 907-919. doi: 10.1111/ epi.13377
17. Chassoux F, Artiges E, Semah F, Laurent A, Landre E et al. (18) F-FDG-PET patterns of surgical success and failure in mesial temporal lobe epilepsy. Neurology 2017; 88 (11): 1045-1053. doi: 10.1212/WNL.0000000000003714
18. Wong CH, Bleasel A, Wen L, Eberl S, Byth K et al. Relationship between preoperative hypometabolism and surgical outcome in neocortical epilepsy surgery. Epilepsia 2012; 53 (8): 1333- 1340. doi: 10.1111/j.1528-1167.2012.03547.x
19. Nelissen N, Van Paesschen W, Baete K, Van Laere K, Palmini A et al. Correlations of interictal FDG-PET metabolism and ictal SPECT perfusion changes in human temporal lobe epilepsy with hippocampal sclerosis. Neuroimage 2006; 32 (2): 684-695. doi: 10.1016/j.neuroimage.2006.04.185
20. Kamm J, Boles Ponto LL, Manzel K, Gaasedelen OJ, Nagahama Y et al. Temporal lobe asymmetry in FDG-PET uptake predicts neuropsychological and seizure outcomes after temporal lobectomy. Epilepsy & Behavior 2018; 78: 62-67. doi: 10.1016/j. yebeh.2017.10.006
21. Nensa F, Demircioglu A, Rischpler C. Artificial intelligence in nuclear medicine. Journal of Nuclear Medicine 2019; 60 (Suppl 2): 29S-37S. doi: 10.2967/jnumed.118.220590
22. Hornik K, Buchta C, Zeileis A. Open-source machine learning: R meets Weka. Computational Statistics 2009; 24 (2): 225-232. doi: 10.1007/s00180-008-0119-7
23. Landwehr N, Hall M, Frank E. Logistic model trees. Machine Learning 2005; 59: 161-205. doi: 10.1007/s10994-005-0466-3
24. Lee JS, Lee DS, Kim SK, Lee SK, Chung JK et al. Localization of epileptogenic zones in F-18 FDG brain PET of patients with temporal lobe epilepsy using artificial neural network. IEEE Transactions on Medical Imaging 2000; 19 (4): 347-355. doi: 10.1109/42.848185
25. Peter J, Khosravi M, Werner TJ, Alavi A. Global temporal lobe asymmetry as a semi-quantitative imaging biomarker for temporal lobe epilepsy lateralization: A machine learning classification study. Hellenic Journal of Nuclear Medicine 2018; 21 (2): 95-101. doi: 10.1967/s002449910800
26. Wieser HG, Blume WT, Fish D, Goldensohn E, Hufnagel A et al. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001; 42 (2): 282-286. doi: 10.1046/j.1528- 1157.2001.4220282.x
27. Ashburner J. Computational anatomy with the SPM software. Magnetic Resonance Imaging 2009; 27 (8): 1163-1174. doi: 10.1016/j.mri.2009.01.006
28. Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage 2003; 19 (3): 1233-1239. doi: 10.1016/S1053-8119(03)00169-1
29. Lancaster JL, Woldorff MG, Parsons LM, Liotti M, Freitas CS et al. Automated Talairach atlas labels for functional brain mapping. Human Brain Mapping 2000; 10 (3): 120- 131. doi: 10.1002/1097-0193(200007)10:3<120::AIDHBM30>3.0.CO;2-8
30. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 2002; 15 (1): 273-289. doi: 10.1006/nimg.2001.0978
31. Bramer M. Principles of data mining. 2nd ed. London, UK: Springer-Verlag; 2013.
32. Özkan Y. Veri Madenciliği Yöntemleri. 4. baskı, İstanbul, Türkiye: Papatya Yayınları; 2020 (in Turkish).
33. Quinlan R. C4.5: Programs for Machine Learning. San Mateo, CA, USA: Morgan Kaufmann Publishers; 1993.
34. Perani D, Della Rosa PA, Cerami C, Gallivanone F, Fallanca F et al. Validation of an optimized SPM procedure for FDGPET in dementia diagnosis in a clinical setting. NeuroImage: Clinical 2014; 6: 445-454. doi: 10.1016/j.nicl.2014.10.009
35. Mendes Coelho VC, Morita ME, Amorim BJ, Ramos CD, Yasuda CL et al. Automated Online Quantification Method for F-18-FDG Positron Emission Tomography/CT Improves Detection of the Epileptogenic Zone in Patients with Pharmacoresistant Epilepsy. Frontiers in Neurology 2017; 8: 453. doi: 10.3389/fneur.2017.00453
36. Drzezga A, Arnold S, Minoshima S, Noachtar S, Szecsi J et al. F-18-FDG PET studies in patients with extratemporal and temporal epilepsy: Evaluation of an observer-independent analysis. Journal of Nuclear Medicine 1999; 40 (5): 737-746.
37. Goffin K, Dedeurwaerdere S, Van Laere K, Van Paesschen W. Neuronuclear assessment of patients with epilepsy. Seminars in Nuclear Medicine 2008; 38 (4): 227-239. doi: 10.1053/j. semnuclmed.2008.02.004