The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Artificial neural network for classifying with epilepsy MEG data

뇌전증 환자의 MEG 데이터에 대한 분류를 위한 인공신경망 적용 연구

  • Yujin Han (Department of Statistics, Duksung Women's University) ;
  • Junsik Kim (Department of Statistics, Duksung Women's University) ;
  • Jaehee Kim (Department of Statistics, Duksung Women's University)
  • 한유진 (정보통계학과 덕성여자대학교) ;
  • 김준식 (정보통계학과 덕성여자대학교) ;
  • 김재희 (정보통계학과 덕성여자대학교)
  • Received : 2023.07.19
  • Accepted : 2023.10.21
  • Published : 2024.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study performed a multi-classification task to classify mesial temporal lobe epilepsy with left hippocampal sclerosis patients (left mTLE), mesial temporal lobe epilepsy with right hippocampal sclerosis (right mTLE), and healthy controls (HC) using magnetoencephalography (MEG) data. We applied various artificial neural networks and compared the results. As a result of modeling with convolutional neural networks (CNN), recurrent neural networks (RNN), and graph neural networks (GNN), the average k-fold accuracy was excellent in the order of CNN-based model, GNN-based model, and RNN-based model. The wall time was excellent in the order of RNN-based model, GNN-based model, and CNN-based model. The graph neural network, which shows good figures in accuracy, performance, and time, and has excellent scalability of network data, is the most suitable model for brain research in the future.

본 연구는 좌측 해마 경화를 보인 내측두엽 뇌전증(left mTLE, mesial temporal lobe epilepsy with left hippocampal sclerosis) 환자군과 우측 해마 경화를 보인 내측두엽 뇌전증(right mTLE, mesial temporal lobe epilepsy with right hippocampal sclerosis) 환자군 그리고 건강한 대조군(healthy controls; HC)으로부터 측정한 뇌자도(magnetoencephalography; MEG) 데이터로 각 그룹을 분류하는 다중 분류 작업에 다양한 인공신경망을 적용하고 그 결과를 비교해 보고자 하였다. 합성곱 신경망, 순환 신경망 그리고 그래프 신경망으로 모델링한 결과, k-fold 정확도 평균은 합성곱 신경망 기반 모델, 그래프 신경망 기반 모델, 순환 신경망 기반 모델 순으로 우수하였다. 또한, 수행 시간은 순환 신경망 기반 모델, 그래프 신경망 기반 모델, 합성곱 신경망 기반 모델 순으로 우수하였다. 정확도 성능과 시간 면에서 모두 좋은 수치를 보이며, 네트워크 데이터의 확장성이 뛰어난 그래프 신경망이 앞으로 뇌 연구에 활용되기 적합한 모델임을 강조하고자 한다.

Keywords

Acknowledgement

본 연구에 MEG 데이터를 제공해주신 서울대 의대 신경외과 정 천기 교수님께 깊이 감사드립니다.

References

  1. Agarwal C, Queen O, Lakkaraju H, and Zitnik M (2023). Evaluating explainability for graph neural networks, Scientific Data, 10, 144.
  2. Aoe J, Fukuma R, Yanagisawa T et al. (2019). Automatic diagnosis of neurological diseases using MEG signals with a deep neural network, Scientific Reports, 9, 5057.
  3. Cho K, Van Merrienboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, and Bengio Y (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation, In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Mexico, 1724-1734.
  4. Demir A, Koike-Akino T, Wang Y, Haruna M, and Erdogmus D (2021). EEG-GNN: Graph neural networks for classification of electroencephalogram (EEG) signals, In Proceedings of 2021 43rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Guadalajara, Mexico (virtual), 1061-1067.
  5. Fan W, Ma Y, Li Q, He Y, Zhao E, Tang J, and Yin D (2019). Graph neural networks for social recommendation, The World Wide Web Conference, 417-426.
  6. Freeman LC (1977). A set of measures of centrality based on betweenness, Sociometry, 40, 35-41. https://doi.org/10.2307/3033543
  7. Freeman LC (1978). Centrality in social networks conceptual clarification, Social Networks, 1, 215-239. https://doi.org/10.1016/0378-8733(78)90021-7
  8. Gadgil S, Zhao Q, Pfefferbaum A, Sullivan EV, Adeli E, and Pohl KM (2020). Spatio-temporal graph convolution for resting-state fMRI analysis. In Medical Image Computing and Computer Assisted Intervention-MICCAI 2020 (pp. 528-538), Springer, Cham, Available from : doi: 10.1007/978-3-030-59728-3 52
  9. Giovannetti A, Susi G, Casti P, Mencattini A, Pusil S, Lopez ME, Di Natale C, and Martinelli E (2021). Deep-MEG: Spatiotemporal CNN features and multiband ensemble classification for predicting the early signs of Alzheimer's disease with magnetoencephalography, Neural Computing and Applications, 33, 14651-14667. https://doi.org/10.1007/s00521-021-06105-4
  10. Gurney K (1997). An introduction to neural networks (pp. 13-16), UCL press, London.
  11. Han YJ and Kim JH (2023). Comparison of TERGM and SAOM: Statistical analysis of student network data, The Korean Journal of Applied Statistics, 36, 1-19. https://doi.org/10.5351/KJAS.2023.36.1.001
  12. Hochreiter S and Schmidhuber J (1997). Long short-term memory, Neural Computation, 9, 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  13. Huang SG, Xia J, Xu L, and Qiu A (2022). Spatio-temporal directed acyclic graph learning with attention mechanisms on brain functional time series and connectivity, Medical Image Analysis, 77, 102370.
  14. Jin SH and Chung CK (2017). Electrophysiological resting-state biomarker for diagnosing mesial temporal lobe epilepsy with hippocampal sclerosis, Epilepsy Research, 129, 138-145. https://doi.org/10.1016/j.eplepsyres.2016.11.018
  15. Jin SH, Jeong W, and Chung CK (2015). Mesial temporal lobe epilepsy with hippocampal sclerosis is a network disorder with altered cortical hubs, Epilepsia, 56, 772-779. https://doi.org/10.1111/epi.12966
  16. Jukic S, Saracevic M, Subasi A, and Kevric J (2020). Comparison of ensemble machine learning methods for automated classification of focal and non-focal epileptic EEG signals, Mathematics, 8, 1481.
  17. Kostas D, Pang EW, and Rudzicz F (2019). Machine learning for MEG during speech tasks, Scientific Reports, 9, 1-13. https://doi.org/10.1038/s41598-019-38612-9
  18. Krizhevsky A, Sutskever I, and Hinton GE (2017). Imagenet classification with deep convolutional neural networks, Communications of the ACM, 60, 84-90. https://doi.org/10.1145/3065386
  19. LeCun Y, Boser B, Denker JS, Henderson D, Howard RE, Hubbard W, and Jackel LD (1989). Backpropagation applied to handwritten zip code recognition, Neural Computation, 1, 541-551. https://doi.org/10.1162/neco.1989.1.4.541
  20. LeCun Y, Bottou L, Bengio Y, and Haffner P (1998). Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86, 2278-2324. https://doi.org/10.1109/5.726791
  21. Li Z, Hwang K, Li K, Wu J, and Ji T (2022). Graph-generative neural network for EEG-based epileptic seizure detection via discovery of dynamic brain functional connectivity, Scientific Reports, 12, 18998.
  22. Mao WL, Fathurrahman HIK, Lee Y, and Chang TW (2020). EEG dataset classification using CNN method, Journal of Physics: Conference Series, 1456, 012017, Available from: doi: 10.1088/1742-6596/1456/1/012017
  23. Memarian N, Kim S, Dewar S, Engel J, and Staba RJ (2015). Multimodal data and machine learning for surgery outcome prediction in complicated cases of mesial temporal lobe epilepsy, Computers in Biology and Medicine, 64, 67-78. https://doi.org/10.1016/j.compbiomed.2015.06.008
  24. Nissen IA, Stam CJ, Van Straaten ECW et al. (2018). Localization of the epileptogenic zone using interictal MEG and machine learning in a large cohort of drug-resistant epilepsy patients, Frontiers in Neurology, 9, 647.
  25. Orchard ER, Ward PG, Chopra S, Storey E, Egan GF, and Jamadar SD (2021). Neuroprotective effects of motherhood on brain function in late life: A resting-state fMRI study, Cerebral Cortex, 31, 1270-1283. https://doi.org/10.1093/cercor/bhaa293
  26. Pantazis D and Adler A (2021). MEG source localization via deep learning, Sensors, 21, 4278.
  27. Petrosian AA, Prokhorov DV, Lajara-Nanson W, and Schiffer RB (2001). Recurrent neural network-based approach for early recognition of Alzheimer's disease in EEG, Clinical Neurophysiology, 112, 1378-1387. https://doi.org/10.1016/S1388-2457(01)00579-X
  28. Qiu J, Tang J, Ma H, Dong Y, Wang K, and Tang J (2018). Deepinf: Social influence prediction with deep learning, In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, 2110-2119.
  29. Roy S, Kiral-Kornek I, and Harrer S (2019). ChronoNet: A deep recurrent neural network for abnormal EEG identification, Artificial Intelligence in Medicine,(pp. 47-56), Springer, Cham. doi: 10.1007/978-3-030-21642-9_8
  30. Scarselli F, Gori M, Tsoi AC, Hagenbuchner M, and Monfardini G (2008). The graph neural network model, IEEE Transactions on Neural Networks, 20, 61-80. https://doi.org/10.1109/TNN.2008.2005605
  31. Song K, Park H, Lee J, Kim A, and Jung J (2023). COVID-19 infection inference with graph neural networks, Scientific Reports, 13, 11469.
  32. Tang X, Liu Y, Shah N, Shi X, Mitra P, and Wang S (2020). Knowing your fate: Friendship, action and temporal explanations for user engagement prediction on social apps, In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, CA, US, 2269-2279.
  33. Thompson WH, Brantefors P, and Fransson P (2017). From static to temporal network theory: Applications to functional brain connectivity, Network Neuroscience, 1, 69-99. https://doi.org/10.1162/NETN_a_00011
  34. Tomlinson SB, Porter BE, and Marsh ED (2017). Interictal network synchrony and local heterogeneity predict epilepsy surgery outcome among pediatric patients, Epilepsia, 58, 402-411. https://doi.org/10.1111/epi.13657
  35. Usman SM, Usman M, and Fong S (2017). Epileptic seizures prediction using machine learning methods, Computational and mathematical methods in medicine, 2017.
  36. Wein S, Malloni WM, Tome AM, Frank SM, Henze GI, W ' ust S, Greenlee MW, and Lang EW (2021). A graph neural network framework for causal inference in brain networks, Scientific Reports, 11, 8061.
  37. Zhou M, Tian C, Cao R, Wang B, Niu Y, Hu T, Guo H, and Xiang J (2018). Epileptic seizure detection based on EEG signals and CNN, Frontiers in Neuroinformatics, 12, 95.