Radiomics-Based Machine Learning to Support Visual Assessment for Improved Epilepsy Classification Using 18F-FDG Brain PET

Document Type : Original Paper

Authors

1 Research Center for Molecular and Cellular Imaging, Advanced Medical Technologies and Equipment Institute, Tehran University of Medical Sciences, Tehran, Iran

2 Medical Physics and Biomedical Engineering Dept., Medical School, Tehran University of Medical Sciences

3 Research Center for Nuclear Medicine, Tehran University of Medical Sciences, Tehran, Iran

10.22038/ijmp.2025.83024.2461

Abstract

Introduction: This study explored radiomics-based machine learning (ML) models as complementary tools to visual evaluation for classifying drug-resistant epilepsy patients and healthy controls using 18F-FDG brain Positron Emission Tomography (PET). Because visual interpretation can be subjective and variable, especially for novice readers, objective and reproducible computational methods are needed.
Material and Methods: Twenty-one drug-resistant epilepsy patients and sixteen healthy controls underwent ¹⁸F-FDG brain PET imaging. From contralateral brain regions, 92 radiomics features (first-order statistics and second-order texture matrices) were extracted. Feature selection included Student’s t-test, principal component analysis, and ridge regression. Logistic regression (LR) and support vector machine (SVM) classifiers were trained and evaluated using 10-fold cross-validation and repeated 80/20 train–test splits. A permutation test (n = 1000) assessed whether differences between classifier performances were statistically significant. LR, chosen for its lower computational cost and interpretability, was used for comparison with human visual assessments.
Results: Across six radiomics feature groups, LR models demonstrated strong performance, with mean accuracy of 0.94(0.05), precision 0.96(0.03), recall 0.92(0.10), specificity 0.97(0.02), and AUC 0.98(0.00). SVM models showed similarly high accuracy 0.98(0.01), precision 0.94(0.05), recall 0.96(0.03), specificity 0.98(0.01), and AUC 0.98(0.00). Novice visual assessments had moderate accuracy (0.62 and 0.67), perfect specificity, lower sensitivity (0.60 and 0.65), and AUCs of 0.80 and 0.825. The final LR model achieved a mean AUC of 0.96(0.01).
Conclusion: This hybrid radiomics-visual approach improves classification accuracy in pre-surgical evaluation of drug-resistant epilepsy. By integrating quantitative radiomics with clinical interpretation, the framework reduces variability and improves reliability for less experienced clinicians.

Keywords

Main Subjects

References

  1. Rigney G, Lennon M, Holderrieth P. The use of computational models in the management and prognosis of refractory epilepsy: a critical evaluation. Seizure. 2021;91:132-40.
  2. Milligan TA. Epilepsy: a clinical overview. The American Journal of Medicine. 2021;134(7):840-7.
  3. Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE classification of the epilepsies: Position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512-21.
  4. Perucca E, Perucca P, White HS, Wirrell EC. Drug resistance in epilepsy. The Lancet Neurology. 2023;22(8):723-34.
  5. Mohammadzadeh P, Nazarbaghi S. The prevalence of drug-resistant-epilepsy and its associated factors in patients with epilepsy. Clinical neurology and neurosurgery. 2022;213:107086.
  6. Bernasconi N, Bernasconi A. Imaging the epileptic brain—time for new standards. Nature Reviews Neurology. 2014;10(3):133-4.
  7. Kini LG, Thaker AA, Hadar PN, Shinohara RT, Brown M-G, Dubroff JG, Davis KA. Quantitative [18] FDG PET asymmetry features predict long-term seizure recurrence in refractory epilepsy. Epilepsy & Behavior. 2021;116:107714.
  8. Thijs RD, Surges R, O'Brien TJ, Sander JW. Epilepsy in adults. The Lancet. 2019;393(10172):689-701.
  9. Duncan JS, Winston GP, Koepp MJ, Ourselin S. Brain imaging in the assessment for epilepsy surgery. The Lancet Neurology. 2016;15(4):420-33.
  10. Hainc N, McAndrews MP, Valiante T, Andrade DM, Wennberg R, Krings T. Imaging in medically refractory epilepsy at 3 Tesla: a 13-year tertiary adult epilepsy center experience. Insights into Imaging. 2022;13(1):99.
  11. Steinbrenner M, Duncan JS, Dickson J, Rathore C, Wächter B, Aygun N, et al. Utility of 18F‐fluorodeoxyglucose positron emission tomography in presurgical evaluation of patients with epilepsy: A multicenter study. Epilepsia. 2022;63(5):1238-52.
  12. Avendaño-Estrada A, Velasco F, Velasco AL, Cuellar-Herrera M, Saucedo-Alvarado PE, Marquez-Franco R, et al. Quantitative analysis of [18F] FFMZ and [18F] FDG PET studies in the localization of seizure onset zone in drug-resistant temporal lobe epilepsy. Stereotactic and Functional Neurosurgery. 2019;97(4):232-40.
  13. Wu D, Yang L, Gong G, Zheng Y, Jin C, Qi L, et al. Characterizing the hyper‐and hypometabolism in temporal lobe epilepsy using multivariate machine learning. Journal of Neuroscience Research. 2021;99(11):3035-46.
  14. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, Van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. European journal of cancer. 2012;48(4):441-6.
  15. Fusco R, Granata V, Grazzini G, Pradella S, Borgheresi A, Bruno A, et al. Radiomics in medical imaging: pitfalls and challenges in clinical management. Japanese journal of radiology. 2022;40(9):919-29.
  16. Mo J, Liu Z, Sun K, Ma Y, Hu W, Zhang C, et al. Automated detection of hippocampal sclerosis using clinically empirical and radiomics features. Epilepsia. 2019;60(12):2519-29.
  17. Park JE, Kim HS. Radiomics as a quantitative imaging biomarker: practical considerations and the current standpoint in neuro-oncologic studies. Nuclear medicine and molecular imaging. 2018;52:99-108.
  18. Sone D, Beheshti I. Clinical application of machine learning models for brain imaging in epilepsy: a review. Frontiers in Neuroscience. 2021;15:684825.
  19. Liu Z, Wang Y, Liu X, Du Y, Tang Z, Wang K, et al. Radiomics analysis allows for precise prediction of epilepsy in patients with low-grade gliomas. NeuroImage: Clinical. 2018;19:271-8.
  20. Zwanenburg A, Vallières M, Abdalah MA, Aerts HJ, Andrearczyk V, Apte A, et al. The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology. 2020;295(2):328-38.
  21. Salem N, Hussein S. Data dimensional reduction and principal components analysis. Procedia Computer Science. 2019;163:292-9.
  22. Palo HK, Sahoo S, Subudhi AK. Dimensionality reduction techniques: Principles, benefits, and limitations. Data Analytics in Bioinformatics: A Machine Learning Perspective. 2021:77-107.
  23. Segato A, Marzullo A, Calimeri F, De Momi E. Artificial intelligence for brain diseases: A systematic review. APL Bioengineering. 2020;4(4):041503.
  24. Parmar C, Grossmann P, Bussink J, Lambin P. Machine learning methods for quantitative radiomic biomarkers. Machine learning applications for Radiomics. 2017;5:125.
  25. Marcot BG, Hanea AM. What is an optimal value of k in k-fold cross-validation in discrete Bayesian network analysis? Computational Statistics. 2021;36(3):2009-31.
  26. Guery D, Rheims S. Clinical management of drug resistant epilepsy: a review on current strategies. Neuropsychiatric Disease and Treatment. 2021:2229-42.
  27. Sultana B, Panzini M-A, Carpentier AV, Comtois J, Rioux B, Gore G, et al. Incidence and prevalence of drug-resistant epilepsy: a systematic review and meta-analysis. Neurology. 2021;96(17):805-17.
  28. Burkholder DB, Ritaccio AL, Shin C. Pre‐surgical Evaluation. Epilepsy. 2021:345-65.
  29. Fitsiori A, Hiremath SB, Boto J, Garibotto V, Vargas MI. Morphological and advanced imaging of epilepsy: beyond the basics. Children. 2019;6(3):43.
  30. Guo J, Guo M, Liu R, Kong Y, Hu X, Yao L, et al. Seizure outcome after surgery for refractory epilepsy diagnosed by 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET/MRI): a systematic review and meta-analysis. World Neurosurgery. 2023;173:34-43.
  31. Tomás J, Pittau F, Hammers A, Bouvard S, Picard F, Vargas MI, et al. The predictive value of hypometabolism in focal epilepsy: a prospective study in surgical candidates. European journal of nuclear medicine and molecular imaging. 2019;46:1806-16.
  32. Pittau F, Grouiller F, Spinelli L, Seeck M, Michel CM, Vulliemoz S. The role of functional neuroimaging in pre-surgical epilepsy evaluation. Frontiers in neurology. 2014;5:31.
  33. Peter J, Houshmand S, Werner TJ, Rubello D, Alavi A. Applications of global quantitative18F-FDG-PET analysis in temporal lobe epilepsy. Nuclear Medicine Communications. 2016;37(3):223-30.
  34. Wissel BD, Greiner HM, Glauser TA, Holland‐Bouley KD, Mangano FT, Santel D, et al. Prospective validation of a machine learning model that uses provider notes to identify candidates for resective epilepsy surgery. Epilepsia. 2020;61(1):39-48.
  35. Yuan J, Ran X, Liu K, Yao C, Yao Y, Wu H, Liu Q. Machine learning applications on neuroimaging for diagnosis and prognosis of epilepsy: A review. Journal of neuroscience methods. 2022;368:109441.
  36. Cantor-Rivera D, Khan AR, Goubran M, Mirsattari SM, Peters TM. Detection of temporal lobe epilepsy using support vector machines in multi-parametric quantitative MR imaging. Computerized Medical Imaging and Graphics. 2015;41:14-28.
  37. Memarian N, Kim S, Dewar S, Engel Jr J, Staba RJ. Multimodal data and machine learning for surgery outcome prediction in complicated cases of mesial temporal lobe epilepsy. Computers in biology and medicine. 2015;64:67-78.
  38. Del Gaizo J, Mofrad N, Jensen JH, Clark D, Glenn R, Helpern J, Bonilha L. Using machine learning to classify temporal lobe epilepsy based on diffusion MRI. Brain and behavior. 2017;7(10):e00801.
  39. Hao J, Xie Y, Liu Q, Xu J, Zhang P, editors. Localization of Epileptogenic Zone Based on Radiomics Features of 18F-FDG PET in Patients with Temporal Lobe Epilepsy. 2021 10th International IEEE/EMBS Conference on Neural Engineering (NER); 2021: IEEE.
  40. Liao K, Wu H, Jiang Y, Dong C, Zhou H, Wu B, et al. Machine learning techniques based on 18F-FDG PET radiomics features of temporal regions for the classification of temporal lobe epilepsy patients from healthy controls. Frontiers in Neurology. 2024;15:1377538.
  41. Zhang Y, Yan P, Liang F, Ma C, Liang S, Jiang C. Predictors of epilepsy presentation in unruptured brain arteriovenous malformations: a quantitative evaluation of location and radiomics features on T2-weighted imaging. World Neurosurgery. 2019;125:e1008-e15.
  42. Cheong E-N, Park JE, Jung DE, Shim WH. Extrahippocampal radiomics analysis can potentially identify laterality in patients with MRI-negative temporal lobe epilepsy. Frontiers in neurology. 2021;12:706576.
  43. Gao A, Yang H, Wang Y, Zhao G, Wang C, Wang H, et al. Radiomics for the prediction of epilepsy in patients with frontal glioma. Frontiers in oncology. 2021;11:725926.
Iranian Journal of Medical Physics
Volume 22, Issue 4 - Serial Number 4
July and August 2025
Pages 218-229

Files

History

  • Receive Date: 03 October 2024
  • Revise Date: 30 July 2025
  • Accept Date: 04 August 2025

Share

How to cite

Statistics