MAKİNE ÖĞRENME ALGORİTMALARI KULLANILARAK ALS TEŞHİSİ İÇİN EEG VERİLERİNİN İSTATİSTİKSEL ÖZELLİK ANALİZİ

Abstract

Amyotrofik Lateral Skleroz (ALS), motor nöronların işlev bozukluğundan kaynaklanan ilerleyici bir nörolojik hastalıktır. Klinik bulgular uzun bir süre boyunca analiz edilir, bu nedenle ALS'nin erken teşhisi oldukça zor ve zaman alıcıdır. Hastalık ilerledikçe, hasta önce felç olur ve sonra ölür. Erken teşhis, hastanın yaşam süresini ve yaşam kalitesini iyileştirmek için çok önemlidir. Elektroensefalografi (EEG) sinyalleri, invaziv olmayan yöntemler kullanılarak ve makul bir maliyetle kolayca kaydedilebilir. EEG sinyalleri biyobelirteç olarak umut vaat etmektedir. Bu çalışma, makine öğrenimi ve istatistiksel öznitelik çıkarma yöntemlerini kullanarak ALS teşhisi için EEG tabanlı bir çözüm sunmaktadır. Veri seti 63.648 EEG kaydı içermektedir. Öznitelikler, istatistiksel öznitelik çıkarma yöntemleri kullanılarak çıkarılmıştır. Veri seti, %80 eğitim ve %20 test olmak üzere iki gruba ayrılmıştır. Ardından, Topluluk Öğrenmesi (EL), Destek Vektör Makineleri (SVM), K-En Yakın Komşu (KNN) ve Karar Ağaçları (DT) makine öğrenme yöntemlerinin performansı karşılaştırılmıştır. Sonuçlar, EL algoritmasının en yüksek performansı elde ettiğini göstermiştir. Algoritma %94,46 doğruluk oranı ve %97,13 özgüllük oranı elde etmiştir. Bu sonuçlar ALS teşhisi için umut vericidir.

Keywords

• Amyotrofik Lateral Skleroz
• Topluluk Öğrenmesi
• EEG
• İstatistiksel Öznitelik Çıkarma
• Makine Öğrenmesi

STATISTICAL FEATURES ANALYSIS OF EEG DATA FOR ALS DIAGNOSIS USING MACHINE LEARNING ALGORITHMS

Abstract

Amyotrophic Lateral Sclerosis (ALS) is a progressive neurological disease resulting from the dysfunction of motor neurons, making early diagnosis difficult due to the slow progression of clinical symptoms. Early diagnosis is crucial to improving the patient's lifespan and quality of life. Electroencephalography (EEG) signals can be recorded non-invasively and at low cost, making them promising biomarkers for ALS diagnosis. Despite the promising results reported in previous EEG-based ALS diagnosis studies, many existing approaches suffer from limited generalizability and insufficient methodological clarity. This study presents an EEG-based solution for ALS diagnosis using machine learning and statistical feature extraction based on time-domain descriptors, namely mean, variance, and higher-order moments, which reflect ALS-related neural alterations. The dataset contained 63,648 channel-level EEG samples obtained from multi-channel recordings of ALS patients and healthy controls. Using the hold-out validation method, the dataset was divided into 80% training and 20% testing sets. The performance of Ensemble Learning (EL), Support Vector Machines (SVM), K-Nearest Neighbor (KNN), and Decision Trees (DT) machine learning methods was compared. The results showed that the EL algorithm achieved the highest performance. The EL algorithm achieved an accuracy of 94.46% and a specificity of 97.13%, demonstrating strong diagnostic potential for ALS classification.

Keywords

• Amyotrophic Lateral Sclerosis
• EEG
• Ensemble Learning
• Machine Learning
• Statistical Feature Extraction.

Supporting Institution

No funding

Ethical Statement

This article does not contain any studies with human participants or animals performed by any of the authors

References

1. S. Uğurlu, M. Yorulmaz, and M. Yıldız, “New turkısh vırtual keyboard applıcatıon usıng sımple camera-mounted eyeglasses,” KONJES, vol. 13, no. 3, pp. 910–926, 2025, doi: 10.36306/konjes.1619363.
2. N. Nakamura et al., “Next-generation sequencing of 28 ALS-related genes in a Japanese ALS cohort,” Neurobiology of Aging, vol. 39, pp. 219.e1–219.e8, 2016, doi: 10.1016/j.neurobiolaging.2015.11.030.
3. J. D. Berry et al., “Epidemiology and economic burden of amyotrophic lateral sclerosis in the United States: a literature review,” Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration, vol. 24, no. 5–6, pp. 436–448, Aug. 2023, doi: 10.1080/21678421.2023.2165947.
4. E. Schönfelder, A. Osmanovic, L. H. Müschen, S. Petri, and O. Schreiber-Katz, “Costs of illness in amyotrophic lateral sclerosis (ALS): a cross-sectional survey in Germany,” Orphanet Journal of Rare Diseases, vol. 15, no. 149, 2020, doi: 10.1186/s13023-020-01413-9.
5. K. Achtert and L. Kerkemeyer, “The economic burden of amyotrophic lateral sclerosis: a systematic review,” European Journal of Health Economics, vol. 22, pp. 1151–1166, 2021, doi: 10.1007/s10198-021-01328-7.
6. K. Stenson et al., “Health care resource utilization and costs across stages of amyotrophic lateral sclerosis in the United States,” Journal of Managed Care & Specialty Pharmacy, vol. 30, no. 11, 2024, doi: 10.18553/jmcp.2024.30.11.1239.
7. O. Hardiman et al., “Amyotrophic lateral sclerosis,” Nature Reviews Disease Primers, vol. 3, p. 17071, 2017, doi: 10.1038/nrdp.2017.71.
8. C. Shipe and S. A. Zivković, “Electrodiagnostic evaluation of motor neuron disorders,” Am. J. Electroneurodiagnostic Technol., vol. 44, no. 1, pp. 30–36, Mar. 2004. Available: https://pubmed.ncbi.nlm.nih.gov/15310030/.

9. M. Sonoo, “Electrodiagnosis of ALS” (in Japanese), Rinsho Shinkeigaku, vol. 46, no. 11, pp. 819–821, Nov. 2006. Available: https://pubmed.ncbi.nlm.nih.gov/17432189/.
10. V. Renga, “Brain connectivity and network analysis in amyotrophic lateral sclerosis,” Neurology Research International, vol. 2022, Art. no. 1838682, Feb. 2022, doi: 10.1155/2022/1838682.
11. M. Proudfoot et al., “Altered cortical beta-band oscillations reflect motor system degeneration in amyotrophic lateral sclerosis,” Human Brain Mapping, vol. 38, no. 1, pp. 237–254, 2017, doi: 10.1002/hbm.23355.
12. A. Öter, “Classıfıcatıon of epıleptıc seızure usıng a one-dımensıonal convolutıonal neural network”, KONJES, vol. 13, no. 3, pp. 822–836, 2025, doi: 10.36306/konjes.1495651.
13. O. K. Cura, G. C. Yilmaz, H. S. Türe and A. Akan, "Classification of Dementia EEG Based on Sub-bands Using Time-Frequency Approaches," 2022 30th Signal Processing and Communications Applications Conference (SIU), Safranbolu, Turkey, 2022, pp. 1-4, doi: 10.1109/SIU55565.2022.9864898.
14. S. Q. O. Omar and C. Tepe, “A Review on the Use of Machine Learning to Process EEG Signals”, Bayburt University Journal of Science, vol. 5, no. 1, pp. 124–137, 2022, doi: 10.55117/bufbd.1099025.
15. Ö. Türk, M. Şeker and M. S. Özerdem, “Channel-based detection of schizophrenia from EEG signals using Hilbert transform,” TBV-BBMD, vol. 13, no. 2, pp. 78–86, 2020. [Online]. Available: https://dergipark.org.tr/tr/download/article-file/1392933
16. R. C. Welsh, L. M. Jelsone-Swain, and B. R. Foerster, “The utility of independent component analysis and machine learning in the identification of the amyotrophic lateral sclerosis diseased brain,” Frontiers in Human Neuroscience, vol. 7, p. 251, Jun. 2013, doi: 10.3389/fnhum.2013.00251.
17. X. Liu et al., “VoxelHop: successive subspace learning for ALS disease classification using structural MRI,” arXiv preprint, arXiv:2101.05131, 2021. Available: https://arxiv.org/abs/2101.05131
18. Y. Roy et al., “Deep learning-based electroencephalography analysis: A systematic review,” Journal of Neural Engineering, vol. 16, no. 5, p. 051001, 2019, doi: 10.1088/1741-2552/ab260c.
19. N. Lohia, C. Mathew, and G. Shankel, “Applying transfer learning and existing EEG datasets to identify patients with ALS,” SMU Data Science Review, vol. 8, no. 2, Art. no. 4, 2024. [Online]. Available: https://scholar.smu.edu/datasciencereview/vol8/iss2/4.
20. F. Latifoğlu, F. O. Bulucu, and R. İleri, “Detection of amyotrophic lateral sclerosis disease using variational mode decomposition and convolutional neural networks from event-related potential signals,” Turkish Journal of Electrical Engineering and Computer Sciences, vol. 29, no. 8, pp. 2840–2854, 2021, doi: 10.3906/elk-2105-86.
21. T. H. Le et al., “An EEG & eye-tracking dataset of ALS patients & healthy people during eye-tracking-based spelling system usage,” figshare, Dataset, 2024. doi: 10.6084/m9.figshare.c.6910027.v1.
22. T. D. Ngo et al., “An EEG & eye-tracking dataset of ALS patients and healthy people during eye-tracking-based spelling system usage,” Scientific Data, vol. 11, 2024, doi: 10.1038/s41597-024-03501-y.
23. D. N. Joanes and C. A. Gill, “Comparing measures of sample skewness and kurtosis,” Journal of the Royal Statistical Society: Series D (The Statistician), vol. 47, no. 1, pp. 183–189, 1998, doi: 10.1111/1467-9884.00122.
24. B. Hjorth, “EEG analysis based on time-domain properties,” Electroencephalography and Clinical Neurophysiology, vol. 29, no. 3, pp. 306–310, Sep. 1970, doi: 10.1016/0013-4694(70)90143-4.
25. A. Subasi, “EEG signal classification using wavelet feature extraction and a mixture of expert model,” Expert Systems with Applications, vol. 32, no. 4, pp. 1084–1093, May 2007, doi: 10.1016/j.eswa.2006.02.005.
26. W. Lu and N. Li, “Integrating virtual reality and physiological feedback into architectural education: enhancing technological literacy and experiential learning,” International Journal of Technology and Design Education, 2025, doi: 10.1007/s10798-025-09999-6.
27. O. A. Rosso et al., “Wavelet entropy: a measure of complexity,” Journal of Neuroscience Methods, vol. 105, no. 1, pp. 65–75, Jan. 2001, doi: 10.1016/S0165-0270(00)00356-3.
28. R. Q. Quiroga and S. Panzeri, “Extracting information from neuronal populations: information theory and decoding approaches,” Nature Reviews Neuroscience, vol. 10, pp. 173–185, Mar. 2009, doi: 10.1038/nrn2578.
29. A. O. Tzimourtas et al., “Integration of 24 feature types to accurately detect and predict seizures using scalp EEG signals,” Sensors, vol. 18, no. 5, p. 1372, May 2018, doi: 10.3390/s18051372.
30. E. N. Nacar and B. Erdebilli, “Sales forecasting using machine learning algorithms,” Journal of Industrial Engineering, vol. 32, no. 2, pp. 307–320, 2021, doi: 10.46465/endustrimuhendisligi.811183.
31. E. Namlı, R. Ünlü, and E. Gül, “Comparison of machine learning techniques and linear regression methods in price prediction; a case study on predicting the prices of used vehicles sold in Turkey,” KONJES, vol. 7, no. 4, pp. 806–821, 2019, doi: 10.36306/konjes.654952.
32. K. Turgut, “Sensitive anomaly detection in large-scale data using support vector machines and optimization techniques,” Journal of Computer Science and Technologies, vol. 6, no. 1, pp. 37–43, 2025, doi: 10.54047/bibted.1567073.
33. Y. S. Balcıoğlu and B. Sezen, “A hybrid system with K-nearest neighbor (KNN) and MLP method: Traffic flow forecasting,” Afyon Kocatepe University Journal of Social Sciences, vol. 26, no. 4, pp. 1801–1816, 2024, doi: 10.32709/akusosbil.1255897.
34. S. R. Safavian and D. Landgrebe, “A survey of decision tree classifier methodology,” IEEE Transactionson Systems, Man, and Cybernetics, vol. 21, no. 3, pp. 660–674, May–Jun. 1991, doi: 10.1109/21.97458.
35. W.-Y. Loh, “Classification and regression trees,” WIREs Data Mining and Knowledge Discovery, vol. 1, no. 1, pp. 14–23, Jan. 2011, doi: 10.1002/widm.8.
36. D. Kılınç, “The Effect of Ensemble Learning Models on Turkish Text Classification”, CBUJOS, vol. 12, no. 2, 2016, doi: 10.18466/cbujos.04526.
37. H. Göker, H. I. Bülbül, and E. Irmak, “The estimation of students' academic success by data mining methods,” in Proc. 2013 12th International Conference on Machine Learning and Applications (ICMLA), vol. 2, pp. 535–539, 2013, doi: 10.1109/ICMLA.2013.173.
38. M. Sokolova and G. Lapalme, “A systematic analysis of performance measures for classification tasks,” Information Processing & Management, vol. 45, no. 4, pp. 427–437, 2009, doi: 10.1016/j.ipm.2009.03.002.
39. J. Chicco and G. Jurman, “The Matthews correlation coefficient (MCC) is more informative than Cohen’s kappa and Brier score in binary classification assessment,” IEEE Access, vol. 9, pp. 78368–78381, 2021, doi: 10.1109/ACCESS.2021.3084050.
40. S. Aliakbaryhosseinabadi et al., “Participant-specific classifier tuning increases the performance of hand movement detection from EEG in patients with amyotrophic lateral sclerosis,” Journal of Neural Engineering, vol. 18, no. 5, 2021, doi: 10.1088/1741-2552/ac15e3.
41. M. Vashkevich, A. Petrovsky, and Y. Rushkevich, “Bulbar ALS detection based on analysis of voice perturbation and vibrato,” in 2019 Signal Processing Algorithms, Architectures, Arrangements, and Applications (SPA), Poznan, Poland, 2019, pp. 268–273, doi: 10.23919/SPA.2019.8936657.
42. O. Bai, P. Lin, D. Huang, D. Y. Fei, and M. K. Floeter, “Towards a user-friendly brain–computer interface: Initial tests in ALS and PLS patients,” Clin. Neurophysiol., vol. 121, no. 8, pp. 1293–1303, Aug. 2010. Available: doi: 10.1016/j.clinph.2010.02.157.
43. C. X. Perez-Ortiz, J. L. Gordillo, O. Mendoza-Montoya, J. M. Antelis, R. Caraza, and H. R. Martinez, “Functional connectivity and frequency power alterations during P300 task as a result of amyotrophic lateral sclerosis,” Sensors, vol. 21, no. 20, Art. no. 6801, 2021, doi: 10.3390/s21206801.
44. J. Thome, R. Steinbach, J. Grosskreutz, D. Durstewitz, and G. Koppe, “Classification of amyotrophic lateral sclerosis by brain volume, connectivity, and network dynamics,” Human Brain Mapping, vol. 43, no. 2, pp. 681–699, 2022, doi: 10.1002/hbm.25679.
45. K. Samanta, S. Roy, V. Marchand-Pauvert, S. Dora, S. Duguez, M. Singh, G. Prasad, and S. Bhattacharyya, “An automated detection of amyotrophic lateral sclerosis from resting-state MEG data using 3D deep convolutional neural network,” in Proc. IEEE Int. Conf. Syst., Man, Cybern. (SMC), 2023, pp. 3337–3342, doi: 10.1109/SMC53992.2023.10393987.