A Data-Driven Eeg Framework For Multiclass Dementia Classification Via Spectral, Microstate, And Connectivity Features
DOI:
https://doi.org/10.29284/bsf3af03Keywords:
Alzheimer’s Disease, Electroencephalography (EEG), Machine Learning, Dementia Classification, Resting State.Abstract
Accurate differential screening of Alzheimer’s disease (AD) and frontotemporal dementia (FTD) remains difficult, motivating scalable EEG-based decision support; however, prior work often benchmarks spectral, microstate, and connectivity features separately and uses heuristic dimensionality reduction. Using a public OpenNeuro eyes-closed resting-state EEG dataset (N=88; 10–20 montage; 500 Hz), we built a standardized pipeline including EEGLAB preprocessing (filtering, ICA artifact removal), temporal standardization (60–360 s), multi-domain feature extraction (spectral band power, theta–alpha ratio, aperiodic 1/f exponent, alpha peak frequency; wPLI-based connectivity with graph metrics; four-class microstate parameters), and leakage-controlled dimensionality reduction via PCA with parallel analysis to automatically determine retained components within each training fold. We compared Elastic Net, RBF-SVM, and Random Forest (500 trees) using repeated 5-fold cross-validation over 100 cycles, evaluating classwise precision/recall/F1 and an integrated radar-area index capturing balanced multi-class performance across AD/FTD/controls. Across models, FTD was frequently misclassified as AD, indicating limited intrinsic separability from resting EEG alone. Feature representation was the dominant determinant of performance: microstate features yielded the largest and most balanced radar areas, spectral features were intermediate, and connectivity features were weakest. The proposed workflow provides a reproducible benchmarking template and practical guidance for EEG/BCI-oriented dementia screening systems.
References
1. J. Bang, S. Spina, and B. L. Miller, "Frontotemporal dementia," The Lancet, vol. 386, no. 10004, pp. 1672-1682, 2015, doi: 10.1016/S0140-6736(15)00461-4.
2. R. Cassani, M. Estarellas, R. San-Martin, F. J. Fraga, and T. H. Falk, "Systematic Review on Resting-State EEG for Alzheimer’s Disease Diagnosis and Progression Assessment," Disease Markers, vol. 2018, no. 1, p. 5174815, 2018, doi: https://doi.org/10.1155/2018/5174815.
3. D. Pirrone, E. Weitschek, P. Di Paolo, S. De Salvo, and M. C. De Cola, "EEG Signal Processing and Supervised Machine Learning to Early Diagnose Alzheimer’s Disease," Applied Sciences, vol. 12, no. 11, p. 5413, 2022. [Online]. Available: https://www.mdpi.com/2076-3417/12/11/5413.
4. C. Babiloni et al., "Measures of resting state EEG rhythms for clinical trials in Alzheimer's disease: Recommendations of an expert panel," (in eng), Alzheimers Dement, vol. 17, no. 9, pp. 1528-1553, Sep 2021, doi: 10.1002/alz.12311.
5. N. Huidobro, R. Meza-Andrade, I. Méndez-Balbuena, C. Trenado, M. Tello Bello, and E. Tepichin Rodríguez, "Electroencephalographic Biomarkers for Neuropsychiatric Diseases: The State of the Art," Bioengineering, vol. 12, no. 3, p. 295, 2025. [Online].
6. M. Naderi and A. Jahanian-Najafabadi, "A systematic review of EEG-based machine learning classifications for obsessive-compulsive disorder: current status and future directions," (in eng), BMC Psychiatry, vol. 25, no. 1, p. 854, Sep 9 2025, doi: 10.1186/s12888-025-07296-z.
7. Vallesi et al., "Resting-state EEG spectral and fractal features in dementia with Lewy bodies with and without visual hallucinations," Clinical Neurophysiology, vol. 168, pp. 43-51, 2024/12/01/ 2024, doi: https://doi.org/10.1016/j.clinph.2024.10.004.
8. K.-I. Jang, Y. I. Kim, H. J. Ju, S. J. An, and P. W. Park, "Dementia classification using two-channel electroencephalography features," Sci Rep-Uk, vol. 15, no. 1, p. 11513, 2025/04/03 2025, doi: 10.1038/s41598-025-93513-4.
9. T. S. Warren, S. A. Shende, J. Ashrafi, G. M. Clements, and R. A. Mudar, "Resting-State EEG Power and Aperiodic Activity in Individuals with Mild Cognitive Impairment and Cognitively Healthy Controls," (in eng), Brain Sci, vol. 15, no. 12, Dec 3 2025, doi: 10.3390/brainsci15121305.
10. P. Martínez-Cañada, E. Perez-Valero, J. Minguillon, F. Pelayo, M. A. López-Gordo, and C. Morillas, "Combining aperiodic 1/f slopes and brain simulation: An EEG/MEG proxy marker of excitation/inhibition imbalance in Alzheimer's disease," (in eng), Alzheimers Dement (Amst), vol. 15, no. 3, p. e12477, Jul-Sep 2023, doi: 10.1002/dad2.12477.
11. S. Schopp et al., "EEG alpha peak frequency: cognitive impairment severity marker in isolated REM sleep behavior disorder," npj Parkinson's Disease, vol. 11, no. 1, p. 185, 2025/07/01 2025, doi: 10.1038/s41531-025-01059-z.
12. M. Lassi et al., "Degradation of EEG microstates patterns in subjective cognitive decline and mild cognitive impairment: Early biomarkers along the Alzheimer’s Disease continuum?," NeuroImage: Clinical, vol. 38, p. 103407, 2023/01/01/ 2023, doi: https://doi.org/10.1016/j.nicl.2023.103407.
13. H. Hou, J. Yu, S. Hao, X. Zhang, and D. Zang, "Electroencephalography microstates as biomarkers for screening Alzheimer's disease: Feasibility analysis and a machine learning classification scheme," Journal of Alzheimer’s Disease, vol. 108, no. 2, pp. 693-702, 2025, doi: 10.1177/13872877251379072.
14. Y. Yan et al., "Functional Connectivity Alterations Based on the Weighted Phase Lag Index: An Exploratory Electroencephalography Study on Alzheimer's Disease," (in eng), Curr Alzheimer Res, vol. 18, no. 6, pp. 513-522, 2021, doi: 10.2174/1567205018666211001110824.
15. Z. Deng et al., "Diagnosis of Alzheimer’s Disease Via Machine Learning Approaches with Integrated Resting-State EEG and ERP Characteristics," Cognitive Computation, vol. 17, no. 6, p. 166, 2025/11/17 2025, doi: 10.1007/s12559-025-10528-9.
16. N. Pudjihartono, T. Fadason, A. W. Kempa-Liehr, and J. M. O'Sullivan, "A Review of Feature Selection Methods for Machine Learning-Based Disease Risk Prediction," (in eng), Front Bioinform, vol. 2, p. 927312, 2022, doi: 10.3389/fbinf.2022.927312.
17. D. Iacobucci, A. Ruvio, S. Román, S. Moon, and P. M. Herr, "How many factors in factor analysis? New insights about parallel analysis with confidence intervals," Journal of Business Research, vol. 139, pp. 1026-1043, 2022/02/01/ 2022, doi: https://doi.org/10.1016/j.jbusres.2021.09.015.
18. G. A. Buzzell, Y. Niu, S. Aviyente, and E. Bernat, "A practical introduction to EEG Time-Frequency Principal Components Analysis (TF-PCA)," Developmental Cognitive Neuroscience, vol. 55, p. 101114, 2022/06/01/ 2022, doi: https://doi.org/10.1016/j.dcn.2022.101114.
19. A. Miltiadous et al. A dataset of EEG recordings from: Alzheimer's disease, Frontotemporal dementia and Healthy subjects, doi: 10.18112/openneuro.ds004504.v1.0.8.
20. A. Delorme and S. Makeig, "EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis," Journal of neuroscience methods, vol. 134, no. 1, pp. 9-21, Mar 15 2004, doi: 10.1016/j.jneumeth.2003.10.009.
21. F. Zhang, F. Wang, L. Yue, H. Zhang, W. Peng, and L. Hu, "Cross-Species Investigation on Resting State Electroencephalogram," Brain Topography, vol. 32, no. 5, pp. 808-824, 2019/09/01 2019, doi: 10.1007/s10548-019-00723-x.
22. Z. Chen, Y. Xie, and C. Li, "Longitudinal Analysis of EEG Microstate Alterations Induced by Cumulative Fatigue from Consecutive Day-Long Tasks," presented at the Proceedings of the 9th International Conference on Electronic Information Technology and Computer Engineering, 2025. [Online]. Available: https://doi.org/10.1145/3766671.3766720.
23. L. Tait et al., "EEG microstate complexity for aiding early diagnosis of Alzheimer’s disease," Sci Rep-Uk, vol. 10, no. 1, p. 17627, 2020/10/19 2020, doi: 10.1038/s41598-020-74790-7.
24. J. L. Horn, "A rationale and test for the number of factors in factor analysis," Psychometrika, vol. 30, no. 2, pp. 179-185, 1965/06/01 1965, doi: 10.1007/BF02289447.
25. H. Lee and H. Cham, "Comparing Accuracy of Parallel Analysis and Fit Statistics for Estimating the Number of Factors With Ordered Categorical Data in Exploratory Factor Analysis," Educational and Psychological Measurement, vol. 84, no. 6, pp. 1173-1202, 2024, doi: 10.1177/00131644241240435.
26. I. O. Woollacott and J. D. Rohrer, "The clinical spectrum of sporadic and familial forms of frontotemporal dementia," (in eng), J Neurochem, vol. 138 Suppl 1, pp. 6-31, Aug 2016, doi: 10.1111/jnc.13654.
27. D. C. Perry et al., "Clinicopathological correlations in behavioural variant frontotemporal dementia," (in eng), Brain, vol. 140, no. 12, pp. 3329-3345, Dec 1 2017, doi: 10.1093/brain/awx254.
28. R. Ossenkoppele et al., "Prevalence of amyloid PET positivity in dementia syndromes: a meta-analysis," (in eng), Jama, vol. 313, no. 19, pp. 1939-49, May 19 2015, doi: 10.1001/jama.2015.4669.
29. Y. Özbek, E. Fide, and G. G. Yener, "Resting-state EEG alpha/theta power ratio discriminates early-onset Alzheimer's disease from healthy controls," (in eng), Clin Neurophysiol, vol. 132, no. 9, pp. 2019-2031, Sep 2021, doi: 10.1016/j.clinph.2021.05.012.
30. Y. Pan, L. Zhu, and Z. L. Wang, "EEG spectral power differences in Alzheimer's disease and frontotemporal dementia," (in eng), Int J Neurosci, pp. 1-9, Nov 21 2025, doi: 10.1080/00207454.2025.2591129.
31. R. Nardone et al., "Usefulness of EEG Techniques in Distinguishing Frontotemporal Dementia from Alzheimer's Disease and Other Dementias," (in eng), Dis Markers, vol. 2018, p. 6581490, 2018, doi: 10.1155/2018/6581490.
32. K. Rascovsky et al., "Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia," (in eng), Brain, vol. 134, no. Pt 9, pp. 2456-77, Sep 2011, doi: 10.1093/brain/awr179.
33. C. M. Michel and T. Koenig, "EEG microstates as a tool for studying the temporal dynamics of whole-brain neuronal networks: A review," (in eng), Neuroimage, vol. 180, no. Pt B, pp. 577-593, Oct 15 2018, doi: 10.1016/j.neuroimage.2017.11.062.
34. A. Khanna, A. Pascual-Leone, C. M. Michel, and F. Farzan, "Microstates in resting-state EEG: current status and future directions," (in eng), Neurosci Biobehav Rev, vol. 49, pp. 105-13, Feb 2015, doi: 10.1016/j.neubiorev.2014.12.010.
35. K. Nishida et al., "EEG microstates associated with salience and frontoparietal networks in frontotemporal dementia, schizophrenia and Alzheimer's disease," (in eng), Clin Neurophysiol, vol. 124, no. 6, pp. 1106-14, Jun 2013, doi: 10.1016/j.clinph.2013.01.005.
36. L. Tait et al., "EEG microstate complexity for aiding early diagnosis of Alzheimer's disease," (in eng), Sci Rep, vol. 10, no. 1, p. 17627, Oct 19 2020, doi: 10.1038/s41598-020-74790-7.
37. M. Vinck, R. Oostenveld, M. van Wingerden, F. Battaglia, and C. M. Pennartz, "An improved index of phase-synchronization for electrophysiological data in the presence of volume-conduction, noise and sample-size bias," (in eng), Neuroimage, vol. 55, no. 4, pp. 1548-65, Apr 15 2011, doi: 10.1016/j.neuroimage.2011.01.055.
38. A. M. Bastos and J. M. Schoffelen, "A Tutorial Review of Functional Connectivity Analysis Methods and Their Interpretational Pitfalls," (in eng), Front Syst Neurosci, vol. 9, p. 175, 2015, doi: 10.3389/fnsys.2015.00175.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
