Title: Machine Learning based Detection of Parkinson’s Disease Using Beta-band PLV and Coherence Measures from EEG

Abstract: Parkinson’s disease (PD) is a progressive neurode generative disorder that impairs motor functions and disrupts functional brain connectivity. Electroencephalography (EEG), with its high temporal resolution, enables investigation of these disruptions in resting-state brain activity. This study aims to differentiate PD patients from healthy controls by analyzing beta band (12–35 Hz) EEG connectivity through two widely used syn chronization metrics: Phase Locking Value (PLV) and magnitude squared coherence. We utilized an open-access EEG dataset from the University of Iowa, comprising 13 individuals with PD and 13 age-matched healthy controls, recorded during eyes open resting state. EEG signals were filtered in the beta band, and both PLV and coherence metrics were calculated across 63 channels. Connectivity heatmaps were generated for whole-brain and motor-related regions (F3, C3, Cz, C4, F4, FC3, FC4). While PD subjects showed increased phase synchronization (PLV), they demonstrated reduced coherence values compared to controls, suggesting abnormal hypersynchrony alongside decreased linear coupling. Statistical comparisons using independent t-tests re vealed significant differences in specific connections (e.g., Cz–F4, F3–C3), particularly within motor areas. These features were used to train six supervised machine learning classifiers including Fine Tree, LDA, Linear SVM, Fine KNN, Naive bayes and neural network. Fine Tree model achieved promising classification accuracy of 84.7%, highlighting the potential of EEG-based features in aiding early diagnosis of Parkinson’s disease. In conclusion, our findings demonstrate that PLV and coher ence in the beta band, especially in motor-related networks, can serve as meaningful biomarkers. Combined with machine learn ing, this approach offers a non-invasive tool for distinguishing PD from healthy controls

Keywords: Parkinson’s disease, EEG, PLV, coherence, machine learning

• A. Oswal, P. Brown, and V. Litvak, “Synchronized neural oscillations and the pathophysiology of Parkinson’s disease,” Brain, vol. 139, no. 2,
• pp. 518–530, 2016.
• A. A. Kühn et al., “Pathological synchronisation in the subthalamic nucleus of patients with Parkinson’s disease relates to both bradykinesia and rigidity,” Experimental Neurology, vol. 215, no. 2, pp. 380–387, 2009.
• A. Singh et al., “Frontal theta and beta oscillations during lower-limb movement in Parkinson’s disease,” Clinical Neurophysiology, vol. 131, no. 3, pp. 694–702, 2020.
• J.-P. Lachaux, E. Rodriguez, J. Martinerie, and F. J. Varela, “Measuring phase synchrony in brain signals,” Human Brain Mapping, vol. 8, no. 4,
• pp. 194–208, 1999.
• S. Aydore, M. Pantazis, and R. M. Leahy, “A note on the phase locking value and its properties,” NeuroImage, vol. 74, pp. 231–244, 2013.
• C. de Hemptinne et al., “Therapeutic deep brain stimulation reduces cortical phase-amplitude coupling in Parkinson’s disease,” Nature Neu- roscience, vol. 18, no. 5, pp. 779–786, 2015.
• A. Mishra et al., “A machine learning approach for detection of Parkin- son’s disease using phase locking value and EEG signals,” Biomedical Signal Processing and Control, vol. 65, Art. no. 102350, 2021.
• M. E. Bastos and P. Fries, “Rhythms for cognition: Communication through coherence,” Neuron, vol. 88, no. 1, pp. 220–235, 2015.
• B. C. M. van Wijk, J. D. Beudel, M. J. J. van den Heuvel, C. M. J. van Laar, and C. F. van Putten, “Low-beta cortico-pallidal coherence de- creases during movement in Parkinson’s disease,” Journal of Physiology, vol. 594, no. 5, pp. 1539–1551, 2016.
• M. Olde Dubbelink et al., “Disrupted brain network topology in Parkin- son’s disease: A longitudinal magnetoencephalography study,” Brain, vol. 137, no. 1, pp. 197–207, 2014.
• M. Y. Özdemir and Y. İşler, “Nonlinear feature analysis for EEG-based biometric authentication via machine learning,” Journal of Intelligent Systems with Applications, vol. 6, no. 2, pp. 27–33, 2023.
• A. Bulut, G. Ozturk, and I. Kaya, “Classification of sleep stages via machine learning algorithms,” Journal of Intelligent Systems with Applications, vol. 5, no. 1, pp. 66–70, 2022.
• M. Degirmenci, Y. K. Yuce, and Y. Isler, “Motor imaginary task classification using statistically significant time domain and frequency domain EEG features,” Journal of Intelligent Systems with Applications, vol. 5, no. 1, pp. 49–54, 2022.
• M. Degirmenci, Y. K. Yuce, and Y. Isler, “Classification of multi-class motor imaginary tasks using Poincare measurements extracted from EEG signals,” Journal of Intelligent Systems with Applications, vol. 5, no. 2, pp. 74–78, 2022.
• M. Degirmenci, Y. K. Yuce, and Y. Isler, “Classification of finger movements from statistically-significant time-domain EEG features,” Frontiers in Human Neuroscience, vol. 18, article 1525139, 2024.
• M. Degirmenci, Y. K. Yuce, M. Perc, and Y. Isler, “EEG channel and feature investigation in binary and multiple motor imagery task predictions,” Journal of Intelligent Systems with Applications, vol. 5, no. 1, pp. 49–54, 2022.
• M. Stoffers et al., “Slowing of oscillatory brain activity is a stable characteristic of Parkinson’s disease without dementia,” Brain, vol. 130, no. 7, pp. 1847–1860, 2007.
• J. Pereira, H. X. Xie, D. Cunha, and D. P. C. Vieira, “Parkinson’s disease classification using machine learning and resting-state EEG,” in Proc. IEEE EMBC, Honolulu, HI, USA, pp. 2048–2051, 2018.
• H. Abásolo, R. Hornero, D. García, M. López, and C. Gómez, “Nonlin- ear analysis of EEG background activity in Alzheimer’s disease patients,” Physiological Measurement, vol. 27, no. 4, pp. 373–383, 2006.
• F. Trambaiolli, A. C. Sato, P. S. Balardin, and J. A. P. de Oliveira, “Functional connectivity and machine learning: A review for clinical neuroimaging,” IEEE Reviews in Biomedical Engineering, vol. 13, pp. 135–148, 2020.
• J. Jenkinson and M. Brown, “New insights into beta oscillations in Parkinson’s disease,” Brain, vol. 134, no. 3, pp. 623–636, 2011.
• J. C. Brittain and P. Brown, “Oscillations and the basal ganglia: Motor control and beyond,” NeuroImage, vol. 85, pp. 637–647, 2014.
• A. Little and P. Brown, “The functional role of beta oscillations in Parkinson’s disease,” Lancet Neurology, vol. 13, no. 6, pp. 557–559, 2014.
• D. M. Palva and S. Palva, “Functional connectivity of neuronal oscilla- tions in the human cortex,” NeuroImage, vol. 61, no. 4, pp. 1035–1046, 2012.
• P. Welch, “The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms,” IEEE Trans. Audio Electroacoust., vol. 15, no. 2, pp. 70–73, 1967.
• T. Brunner et al., “Coherence analysis of intracranial EEG signals reveals focal seizure networks,” Clinical Neurophysiology, vol. 125, no. 1, pp. 129–138, 2014.
• D. Bassett and E. Bullmore, “Small-world brain networks,” The Neu- roscientist, vol. 12, no. 6, pp. 512–523, 2006.
• B. H. Fleisher et al., “Deep learning-based EEG classification system for Parkinson’s disease,” Scientific Reports, vol. 12, pp. XX–XX, 2022.
• T. Roy, A. Garg, and P. Prasad, “A hybrid deep learning model for detection of Parkinson’s disease using EEG signals,” IEEE Access, vol. 9, pp. 103960–103969, 2021.
• M. F. Anjum, S. Dasgupta, R. Mudumbai, A. Singh, J. F. Cavanagh, and N. S. Narayanan, “Linear predictive coding distinguishes spectral EEG features of Parkinson’s disease,” Parkinsonism Relat. Disord., vol. 79, pp. 79–85, 2020.