Next Article in Journal
From Prediction to Planning: A Spectral-Temporal GNN and Bi-Directional Decoding RL Framework
Previous Article in Journal
High-Frequency Infrared Thermography Reveals Short-Term Pressure Variations in CO2 Natural Vents at Mefite d’Ansanto (Italy)
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Signals
Volume 7
Issue 3
10.3390/signals7030046
signals-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exploratory Analysis of Electroencephalography Characteristics Shared by Major Depressive Disorder and Parkinson’s Disease: A Database Study

by
Chia-Yen Yang
1,*,
Fan-Ning Kuo
2 and
Hsin-Yung Chen
3
1
Department of Biomedical Engineering, Chung Yuan Christian University, Taoyuan 320314, Taiwan
2
Institute of Biomedical Engineering, National Yang Ming Chiao Tung University, Taipei 112304, Taiwan
3
Department of Occupational Therapy, College of Medicine, I-Shou University, Kaohsiung 824005, Taiwan
*
Author to whom correspondence should be addressed.
Signals 2026, 7(3), 46; https://doi.org/10.3390/signals7030046
Submission received: 10 March 2026 / Revised: 20 April 2026 / Accepted: 23 April 2026 / Published: 8 May 2026
Browse Figures
Versions Notes

Abstract

Despite being distinct clinical entities, major depressive disorder (MDD) and Parkinson’s disease (PD) have some shared physiological pathways, including mitochondrial dysfunction and inflammation. Our interest was whether these common physiological mechanisms are reflected in brain activity variations as well. Therefore, this study aimed to identify common characteristics in resting-state electroencephalography (EEG) between the conditions by comparing features among patients with MDD, PD, and healthy controls. The methodology comprised two stages: analyzing differences between patients and healthy individuals and exploring consistent trends between PD and MDD, based on EEG data from PRED + CT database. Age-corrected regression analysis of five EEG features revealed PD and MDD had the following overlapping features: shared abnormalities in theta, alpha and beta relative power, as well as sample entropy in the delta (centroparietal, temporal, and parietal areas), theta (parieto-occipital), and gamma (central) bands. Furthermore, interhemispheric asymmetry was evident across all bands, especially in the frontal and centroparietal regions. When combining these findings with their directional trends (positive or negative), common EEG features included increased theta and decreased alpha-beta power, along with increased parieto-occipital and reduced gamma entropy at FCz. These findings suggest shared EEG markers between PD and MDD, supporting the potential for efficient neurological disorder diagnosis.

1. Introduction

1.1. Parkinson’s Disease

Parkinson’s disease (PD) is the second most common neurodegenerative disorder. The economic burden of PD is projected to reach approximately US$79.1 billion in the United States by 2037 [1]. The motor symptoms of PD include tremors, muscle rigidity, and bradykinesia, whereas the nonmotor symptoms include depression, sleep disturbances, and cognitive impairment. Notably, several studies have indicated that certain nonmotor symptoms can manifest years before the onset of motor symptoms [2,3].
In research on nonmotor symptoms, the commonly employed functional neuroimaging tools include functional magnetic resonance imaging, positron emission tomography, and electroencephalography (EEG). For example, Yuvaraj et al. [4] employed EEG to examine changes in brain activity during emotional processing in patients with PD and reported increased theta and gamma power and decreased alpha and beta power during negative emotional experiences compared with healthy controls. In patients with PD, Iyer et al. [5] investigated the associations of depression, anxiety, and cognitive impairment with brain functional connectivity. Their findings revealed that, during the resting state, patients exhibited increased connectivity in the theta band between the frontoparietal lobes and in the gamma band between the frontoparietal and frontotemporal lobes, with this connectivity diminishing during cognitive tasks. Increased theta band connectivity in the frontoparietal lobe and gamma band connectivity in the frontotemporal lobe were associated with cognitive impairment, whereas decreased connectivity in these bands was correlated with higher depression score and longer PD duration. Furthermore, Pappalettera et al. [6] investigated the complexity of brain waves in patients with PD and reported higher approximate entropy values in all brain regions compared with those for healthy controls (HCs), indicating greater complexity of brain activity. Their findings imply that PD may induce disruptions in the brain’s overall structure and function rather than being limited to specific brain regions.

1.2. Depressive Disorder

The World Health Organization has predicted that depression will become the leading cause of global socioeconomic burden by 2030 [7]. Current practices for diagnosing major depressive disorder (MDD) predominantly rely on subjective evaluations, including clinical consultations and the use of standardized assessment tools such as the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) and Beck Depression Inventory (BDI) [8].
Given the lack of precise objective clinical markers for diagnosing depression, numerous researchers have employed EEG to investigate brain abnormalities associated with the disorder. Mumtaz et al. [9] employed EEG features and machine learning for the automated diagnosis of depression and revealed lower theta and alpha power but higher delta and beta power in patients with depression compared with HCs. Additionally, in terms of alpha hemispheric asymmetry, patients with depression exhibited notably lower alpha power in the right frontal lobe compared with the left. The authors suggested that the observed decline in theta and alpha EEG signals of the patients with depression may reflect brain functional abnormalities associated with the disorder. Liu et al. [10] further investigated brain wave variations in individuals with their eyes open and closed to identify characteristics indicative of depression. Their study revealed significant differences in brain wave patterns, particularly in the beta and gamma frequency bands, indicating that these bands may serve as early physiological markers of depression. Furthermore, Wu et al. [11] examined the spectral characteristics and functional connectivity of resting-state brain waves in individuals with late-life depression. They reported significantly higher alpha and beta power in patients with depression compared with controls, with alpha power elevation primarily in the central parietal lobe and beta power increases in the parietal, central, and occipital lobes. These abnormalities may be indicative of cortical excitability, inhibition, and hyperreactivity. In summary, EEG has proven to be a viable tool for detecting neurological disorders, including depression.

1.3. Connections Between Depression and PD

Although depression and PD are distinct conditions, many studies have identified shared physiological pathways [12,13,14,15]:
  • Mitochondrial dysfunction: Mitochondrial dysfunction can trigger oxidative stress by disrupting the mitochondrial electron transport chain, resulting in overproduction of reactive oxygen species (ROS), including free radicals. Excessive ROS can damage cellular lipids, proteins, and DNA, impairing cellular function and stability. DNA breakage, in particular, is a pivotal factor in the pathogenesis of both depression and PD, serving as a fundamental mechanism underlying the progression of these disorders.
  • Monoamine hypothesis: Decreased levels of monoamine neurotransmitters can impair neuronal communication, leading to compromised brain functions and symptoms characteristic of depression and PD. For instance, decreased serotonin due to neuronal mutations in the midbrain raphe nucleus may affect the striatum and prefrontal cortex, influencing sleep, mood, and appetite regulation. Furthermore, psychological stress can activate the hypothalamic–pituitary–adrenal (HPA) axis, leading to a reduction in dopamine neurons in the substantia nigra pars compacta, which contributes to depression and cognitive impairment in PD. Stress-induced increases in the norepinephrine level in the locus coeruleus may influence brain regions, leading to inattention, fatigue, and abnormalities in the sympathetic nervous system.
  • Inflammation hypothesis: Research on depression has indicated that chronic stress can over-activate the immune system, triggering inflammatory responses, which in turn, disrupt amino acid metabolism, induce neurotoxicity, deplete glial cells, and impair the negative feedback regulation of the HPA axis, causing sickness behavior. Similarly, studies on PD have suggested that dysregulation of the HPA axis can affect glucocorticoid release, increasing the permeability of the blood–brain barrier. This increased permeability allows cytotoxic molecules to penetrate the brain, leading to degeneration of dopamine neurons and the onset of clinical symptoms associated with PD.
It aroused our interest that whether the physiological mechanisms shared by PD and MDD are also reflected in brain potential activity. Hence, in this study, we aimed to explore the differences in resting-state brain waves between patients with PD and HCs as well as between patients with MDD and HCs and further identify characteristics common to the two patient groups. These shared features could be applied in diagnostic assessments for both conditions. Based on previous research, we selected five commonly studied EEG features spanning both the time and frequency domains for analysis: mean frequency, relative power, sample entropy, detrended fluctuation analysis (DFA), and interhemispheric absolute power asymmetry. To determine the shared EEG characteristics of PD and MDD, we performed t-tests to identify significant differences between patients and HCs. By examining overlapping features with consistent trends—whether both increased or both decreased—we aimed to establish commonalities between the two disorders. Future applications may include implementing a cross-disease transfer learning model to address the problem of data scarcity in certain disorders, thereby enhancing the accuracy and efficiency of diagnostic processes across multiple conditions.

2. Materials and Methods

2.1. Datasets

All EEG data used in this study were downloaded from the Patient Repository of EEG Data + Computational Tools (PRED + CT) database, a resource developed by Cavanagh et al. [16,17].

2.1.1. PD Dataset

The study included 20 patients with PD and 20 HCs, aged 45 to 80 years with Mini-Mental State Examination scores above 26, from the database (Table 1). PD severity was assessed using the Unified Parkinson’s Disease Rating Scale (UPDRS) motor scores by a neurologist.
Participants were instructed to minimize movements and think of nothing for 3 min with eyes closed, avoiding sleep. EEG signals were recorded with a 64-channel Brain Vision system (Brain Products Ltd., Munich, Germany), digitized at 500 Hz, and filtered between 0.1 and 100 Hz. CPz was used as the online reference. Only data from participants who were off medication during the experiment were used for subsequent analysis.

2.1.2. MDD Dataset

The study included 20 MDD patients and 20 HCs aged 18 to 25 from the database. HCs had a BDI score of less than 7, while MDD patients had a score above 13 and were diagnosed via the Structured Clinical Interview for Depression (Table 1).
Participants were instructed to minimize movements and think of nothing for 5 min with eyes closed, avoiding sleep. EEG signals were recorded with a 64-channel SynAmps2 system (Neuroscan, Charlotte, NC, USA), digitized at 500 Hz, and filtered between 0.5 and 100 Hz, with an online reference between Cz and CPz, and electrode impedance below 10 kΩ. Regarding medication status, the MDD dataset consisted of individuals with no current psychoactive medication use at the time of the experiment.

2.2. EEG Processing

The EEG signals were obtained from the database in their raw form, and all subsequent preprocessing, feature extraction, and statistical analyses were performed by the authors. All processing steps were executed using the MNE and SciPy toolboxes (versions 1.8.0 and 1.13.1, respectively) and custom Python scripts (version 3.9.13). Pre-processing of EEG signals involved six stages: downsampling the data to 250 Hz, segmenting the eyes-closed data, detrending to eliminate means, offsets, and slow linear drifts, bandpass filtering between 0.5 and 50 Hz, re-referencing to the average, and selecting the 56 overlapping channels from the two datasets, including Fp1, Fp2, AF3, AF4, F1, F2, F3, F4, F5, F6, F7, F8, Fz, FC1, FC2, FC3, FC4, FC5, FC6, FCz, FT7, FT8, C1, C2, C3, C4, C5, C6, CP1, CP2, CP3, CP4, CP5, CP6, Cz, T7, T8, TP7, TP8, P1, P2, P3, P4, P5, P6, P7, P8, PO3, PO4, PO7, PO8, POz, Pz, O1, O2 and Oz.
After being preprocessed, the EEG signals were decomposed into five frequency bands by using discrete wavelet decomposition: delta (0.5–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), beta (12–30 Hz), and gamma (30–50 Hz). To streamline calculations and ensure consistency across datasets, a 60 s segment, spanning from 10 to 70 s, was extracted from each recording. Features were calculated using a 2 s window without overlap. In line with previous EEG studies on PD and MDD (Table 2), five distinctive features were extracted for analysis. Each feature was subsequently normalized on a channel-wise basis within each database using min–max scaling to minimize systematic bias.
  • Mean frequency: The sum of the product of the spectrogram intensity and the frequency, divided by the total sum of the spectrogram intensity. M F = ( ∑ f = f i r s t e n d F f × f ) / ( ∑ f = f i r s t e n d F f ) , where f is frequency and F(f) is Fourier transformation of signals at that frequency.
  • Frequency power: This is the relative power calculated in each band, and it is divided by the sum of the power across the entire frequency.
  • Sample entropy: Template vectors, Ym(i), constructed using a given embedding dimension m were used to measure the negative natural logarithm of the probability ln(ψm(r)/ψm + 1(r)), where ψm(r) denotes the number of template vectors having a distance, d(Ym(i), Ym(j)) = max|xi + k − xj + k|, smaller than tolerance r.
  • DFA: A cumulative sum, Y(i), calculated from the time-series data set x, where i = 0, 1, …, N, was divided into time windows of length n and was linearly fitted using least-squared errors. The slope of the log of the root-mean-square deviation from the trend, denoted as F n = 1 N ∑ i = 1 N Y i − Y n i 2 , where n = 0.2, 0.25, 0.3, …, 2, against log n within each time window was then calculated using least-squares.
  • Interhemispheric asymmetry: A total of 14 pairs (i.e., FP1-FP2, AF3-AF4, F3-F4, F7-F8, FC1-FC2, FC5-FC6, C3-C4, T7-T8, CP1-CP2, CP5-CP6, P3-P4, P7-P8, PO3-PO4 and O1-O2) indices were calculated as (R − L)/(R + L), where R and L are right and left hemisphere powers, respectively.

2.3. Statistical Analysis

After calculating the features, we performed Spearman’s rank correlation between the BDI or UPDRS scores and each EEG feature within the patient groups (MDD or PD) to verify whether the extracted EEG features were influenced by disease severity. Subsequently, to ensure these correlations were not confounded by aging, a Multiple Linear Regression (MLR) model estimated by Ordinary Least Squares (OLS) was employed to regress out the effect of age. The resulting p-values were adjusted using the False Discovery Rate (FDR) method to account for multiple comparisons across the 56 channels. This analysis aimed to confirm that the observed features represent stable neurophysiological traits of the disorders rather than fluctuations in symptom severity.
To rigorously compare the EEG features between patients (MDD or PD) and HCs while addressing the substantial age differences between datasets, we employed OLS-based regression analysis. The model was defined as: E E G = β 0 + β 1 D i s e a s e + β 2 A g e + ϵ . The p-values associated with the disease effect across all 56 channels were then corrected using FDR (q < 0.05). This approach ensures that the identified differences specifically reflect pathological signatures rather than age-related physiological changes. We then analyzed the overlapping differences between the two disorders. Next, we computed the average feature values for each group by subtracting the healthy group values from the disease group values to assess whether the trends in features with significant differences were similar in the two conditions. For instance, if the result was positive for both the PD and MDD groups compared to the healthy group (or negative for both), the trend was considered similar (Figure 1).

3. Results

Correlation analysis, after controlling for age, showed no significant correlation (p > 0.05, FDR-corrected) between disease severity (BDI or UPDRS scores) and the selected EEG features. Therefore, subsequent analyses were conducted using the unified cohort data for each disorder.

3.1. Differences in EEG Features Between Patient and Healthy Subject Groups

The OLS regression analysis, which accounted for age as a covariate, revealed the five features indicated the following:
  • Mean frequency: MDD patients exhibited significant disease effects (p < 0.05, FDR-corrected) compared with HCs in the fronto-central, central, centroparietal, parietal, and parieto-occipital areas, whereas PD patients showed differences only in the right frontal area. No overlapping significant differences were found between the two groups (Figure 2).
    Signals 07 00046 g002
    Figure 2. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of mean frequency between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD.
    Figure 2. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of mean frequency between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD.
    Signals 07 00046 g002
  • Frequency power: MDD patients exhibited significant disease effects (p < 0.05, FDR-corrected) compared with HCs in the theta and alpha bands across nearly the entire brain. In the delta band, differences were found in the fronto-central, central, centroparietal, and parietal areas, while the beta band showed differences in the frontal, fronto-central and central areas. PD patients also showed significant differences in the theta band across almost the entire brain area. In the delta band, differences were discovered at F5; in the alpha band they were found over the right parietal area; in the beta band, across the frontal, fronto-central, central and parieto-occipital areas; and in the gamma band, at AF4. Notably, the two groups showed overlapping significant differences in the theta band (across nearly the entire brain area), the alpha band (within the right parietal area) and the beta band (across the frontal, fronto-central, central areas) (Figure 3).
    Signals 07 00046 g003
    Figure 3. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of relative power between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD, while red circles indicate electrodes significant for both conditions.
    Figure 3. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of relative power between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD, while red circles indicate electrodes significant for both conditions.
    Signals 07 00046 g003
  • Sample entropy: MDD patients exhibited significant disease effects (p < 0.05, FDR-corrected) compared with HCs in the delta band across nearly the entire brain. In the theta band, differences were discovered in the frontal, parietal, and parieto-occipital areas; in the alpha band, they were found in the central area; in the beta band, over the frontal area; and in the gamma band, across the frontal, central, centroparietal, parietal and parieto-occipital areas. Conversely, PD patients showed significant differences in the delta band within the centroparietal, temporal, and parietal areas. In the theta band, differences were discovered in the frontal and parieto-occipital areas; in the alpha band, they were found in the parietal area; and in the gamma band, over the central area. Furthermore, the two groups showed overlapping significant differences in the delta band (centroparietal, temporal, and parietal areas), the theta band (parieto-occipital area), and the gamma band (central area) (Figure 4).
    Signals 07 00046 g004
    Figure 4. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of sample entropy between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD, while red circles indicate electrodes significant for both conditions.
    Figure 4. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of sample entropy between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD, while red circles indicate electrodes significant for both conditions.
    Signals 07 00046 g004
DFA: MDD patients exhibited significant disease effects (p < 0.05, FDR-corrected) compared with HCs in the delta band across the frontal, central, and left parietal areas. In the theta band, differences were found in the frontal, fronto-central, and parieto-occipital areas, while in the beta band, a difference was found at C5. In contrast, PD patients showed significant differences only in the alpha band over the parieto-occipital area. No overlapping significant differences were found between the two groups (Figure 5).
4.
Interhemispheric asymmetry: Compared with HCs, MDD patients exhibited significant disease effects (p < 0.05, FDR-corrected) in the delta band at AF3-AF4, C1-C2, C3-C4 and TP7-TP8. In the theta band, significant differences were found at F7-F8, FC5-FC6 and C3-C4; in the alpha band, they were found at F1-F2, F5-F6 and FC1-FC2; in the beta band, they were found at AF3-AF4, F1-F2, T7-T8 and TP7-TP8; and in the gamma band, they were found across the frontal, fronto-central, central, and centroparietal areas. Similarly, PD patients showed significant differences in the delta band at F1-F2 and TP7-TP8. In the theta band, a significant difference was found at F7-F8. In the alpha band, significant differences were found at F1-F2, F5-F6 and P5-P6; in the beta band, they were found at AF3-AF4 and F1-F2; and in the gamma band, they were found across the frontal, central, and centroparietal areas. Furthermore, the two groups showed overlapping significant differences in the delta band at TP7-TP8, in the theta at F7-F8, in the alpha at F1-F2 and F5-F6, in the beta band at AF3-AF4 and F1-F2, and in the gamma band at F1-F2, F7-F8, C1-C2, C3-C4, C5-C6, CP3-CP4 and CP5-CP6 (Figure 6).

3.2. Similarity in EEG Features Between PD and MDD Groups

Next, we computed the average values for the five features in each group, subtracting the healthy control values from those of the PD or MDD groups. Significant differences (p < 0.05) with consistent trends (both positive or both negative) prompted further discussion. In both patient groups, the following findings were obtained:
  • Frequency power: Compared with HCs, patients with PD and MDD showed significant increases in the theta band at 31 of 56 electrodes (trend ratio = 0.55), including AF3, AF4, Fp1, Fp2, F1, F3, F4, F6, Fz, FC2, FCz, C1, C2, CP1, CP4, Cz, TP8, P1, P2, P4, P5, P7, P8, Pz, PO3, PO4, PO7, PO8, O1, O2 and Oz (refer to Figure 3). Conversely, significant decreases were observed in the alpha (at PO4, PO7, and PO8) and beta bands (at C2, Fz, and Pz).
  • Sample entropy: Compared with HCs, patients with PD and MDD showed significant increases were observed in the theta band at PO4, PO7 and PO8. Conversely, significant decreases were observed in the gamma band at FCz.

4. Discussion

The use of independent datasets with different acquisition protocols reflects challenges commonly encountered in real-world clinical settings. While such technical heterogeneity may introduce variability, this study represents a preliminary attempt to assess whether shared neurophysiological signatures can be identified across heterogeneous sources. The observation of consistent trends despite these differences suggests a degree of cross-system robustness.

4.1. Differences in EEG Features Between Patient and Healthy Subject Groups

In the present study, both MDD and PD patients exhibited distinct alterations in resting-state EEG features compared with healthy controls. These deviations were particularly evident in theta, alpha and beta relative power, sample entropy in the delta band (centroparietal, temporal, and parietal areas), the theta band (parieto-occipital area), and the gamma band (central area), and interhemispheric asymmetry across specific bands. Together, these findings may reflect dysfunctions in cortical excitability and network regulation. In the context of neurological disorders, cognitive, emotional, and sleep dysfunctions have been identified as major concerns [34]. For example, Bockova et al. [35] identified correlations between nonmotor symptoms of PD and brain oscillatory activities: theta activity was associated with impulse control, alpha activity with depression, beta activity with cognitive impairment, and gamma activity with a loss of cognitive flexibility. Additionally, Grin-Yatsenko et al. [36] examined the EEG spectrum in early-stage depression and revealed increased slow wave activity, indicating decreased cortical activation in parietal and occipital regions, which may be associated with anxiety symptoms. Özçoban and Tan [37] investigated brain activity in 48 MDD patients and 78 healthy controls using EEG power spectral analysis. They found that MDD patients exhibited significantly higher absolute and relative power in the theta and beta bands, along with reduced alpha power, compared to controls. These findings suggest that MDD alters brain mechanisms and cognitive functions, as reflected by changes in theta and alpha power, which are consistent with our results.
Most studies on MDD have primarily focused on frontal alpha asymmetry (FAA), with some associating it with neural system functioning and others proposing it as a potential diagnostic biomarker. Typically, MDD patients show reduced activation in the left frontal cortex relative to the right (i.e., increased alpha power in the left frontal cortex), whereas healthy individuals exhibit the opposite pattern [38]. Such alpha asymmetry has also been reported in neurodegenerative conditions. For example, da Rocha et al. [39] examined oscillatory asymmetries in 37 PD patients and 24 healthy controls, finding that alpha asymmetry may serve as a neurophysiological marker of PD motor symptoms—particularly resting tremors, gait/balance issues, freezing—and of specific cognitive domains such as attention and orientation. However, Özçoban and Tan [37] emphasized that, beyond alpha asymmetry, asymmetries in other frequency bands may also yield valuable insights into the neurophysiological effects. This aligns with previous EEG studies suggesting that the diagnostic value of FAA alone may be limited [40,41].
Further investigations have highlighted the role of EEG complexity measures in understanding depression. Lau et al. [42] reviewed studies employing complexity analysis and reported that patients with depression generally exhibit reduced EEG complexity compared to healthy controls. This reduction has been attributed to impaired emotion regulation and a tendency to remain fixated on negative emotional states, ultimately leading to rigid cognitive patterns. The decreased complexity and increased predictability observed in MDD were reflected in reduced sample entropy and altered DFA values [43]. Similarly, Kanunikov and Kleeva [44] explored the association between EEG entropy indices and depression severity, as assessed by the BDI. They observed that all types of EEG entropy were significantly negatively correlated with BDI scores, suggesting that individuals with higher EEG complexity may be more resilient against the somatic manifestations of depression. Reduced complexity is observed not only in depression but also in PD. For instance, Yi et al. [45] analyzed resting-state EEG from 18 PD patients and 18 age-matched healthy controls to examine early abnormalities in brain activity. Their permutation entropy (PE) analysis revealed significantly lower entropy values in PD patients, suggesting reduced EEG complexity in early PD. Sub-band analyses further showed that gamma, beta, and alpha rhythms in PD patients exhibited lower complexity compared with controls, indicating that PE may be an effective method for detecting abnormal dynamic changes in early PD. The ability of the cortex to sustain complex functions is fundamental to cognitive health, and once compensatory capacity is depleted, cognitive decline may become clinically evident [46].
Nevertheless, while some studies report reduced entropy in PD or MDD, others have found varying or even contradictory trends, as also observed in DFA analyses [47,48]. These discrepancies may reflect compensatory mechanisms or heterogeneity in different disease progression. Overall, these results suggest that EEG complexity measures capture dynamic neural processes that may be overlooked by linear approaches and hold promise as biomarkers for neuropsychiatric pathology, warranting further investigation.

4.2. Similarity in EEG Features Between PD and MDD Groups

Beyond patient–control comparisons, our analysis also explored the overlap with consistent trends between MDD and PD. Several EEG alterations, such as increased theta relative power, decreased alpha and beta relative power, elevated parieto-occipital gamma entropy and reduced mid-frontal (FCz) gamma entropy, were consistently observed in both groups, suggesting potentially shared neurophysiological mechanisms. From a neurotransmitter perspective, dopamine, serotonin, and norepinephrine are considered major focal points, as their abnormalities are associated with α-synuclein, a protein that regulates presynaptic vesicle trafficking and facilitates neurotransmitter release. For example, a review study by Salomon and Cowan [49] suggested that the effects of depression on specific EEG bands may be partially due to abnormal serotonin (5-HT) function. Specifically, lower 5-HT synaptosomal accumulation is associated with increased EEG alertness, and the EEG pattern demonstrates ultradian rhythmic changes in theta frequency (4–8 Hz) of cortical and cortico-thalamic oscillations. This demonstrates the key regulatory role of 5-HT on cortical neural activity and supports a close association between the EEG features of depression and serotonin functional subgroups [50]. Olbrich et al. [51] discovered that alpha-band phase synchronization in untreated patients with MDD was correlated with gamma-aminobutyric acid (GABA) level, indicating that imbalances in GABA and glutamate affect brain phase synchronization, a process also influenced by norepinephrine and serotonin. Moreover, Ip et al. [52] examined the brain arousal response to intravenous ketamine administration in patients with MDD and reported elevated alpha and diminished delta activity in parietal and occipital lobes. This pattern was correlated with increased sympathetic activity, activation of the HPA axis, and shortened REM sleep latency, potentially contributing to difficulties in sleep onset and reduced sleep quality.
Deficits in dopaminergic, serotonergic, and noradrenergic systems may contribute to the impairments in executive, visuospatial, attentional, and memory functions observed in PD [53], as reflected in EEG alterations. For example, Kemp et al. [54] reported that increased theta and delta power and decreased alpha and beta power were inversely correlated with the density of noradrenergic α2 adrenoceptors in the frontal cortex, suggesting that the loss or dysfunction of noradrenergic neurotransmission may underlie cognitive impairment and pathological EEG slowing in PD. Similarly, Schreiner et al. [55] revealed that cognitive impairment in PD is associated with low-frequency brain waves in the frontal lobe, indicating that slow waves may play a role in regulating neurodegeneration by affecting proteins such as α-synuclein, tau, and β-amyloid, thereby influencing the progression of brain lesions. Extending this line of evidence, Waninger et al. [56] explored neurophysiological biomarkers of PD as evaluation indicators for clinical treatment and discovered that increased phosphorylation of α-synuclein was associated with EEG abnormalities related to cognitive decline. These abnormalities were specifically noted in the delta band in the anterior cingulate gyrus, alpha band in the temporal lobe, and beta band in the hippocampus. They further demonstrated that gamma-band coherence strongly correlates with disease severity and dopaminergic tone across all striatal subregions, suggesting that gamma oscillations may serve as a robust neurophysiological biomarker complementary to beta abnormalities in PD.
In summary, these studies have highlighted overlapping physiological mechanisms between PD and MDD, which lead to analogous symptoms in the two conditions. Although some consistencies may arise from distinct neural pathways producing convergent symptomatology, the substantial age discrepancy between cohorts necessitates treating these findings as exploratory. Nevertheless, these shared signatures provide a valuable foundation for further investigation into their broader physiological significance. These findings broaden the understanding of cross-disease neural dynamics and suggest potential avenues for developing shared biomarkers and leveraging transfer learning strategies in clinical research.

4.3. Limitations

This study has some limitations. First, the age range of participants on the two databases was inconsistent. Although we employed OLS regression to statistically control for the confounding effect of age, we recognize that age-related EEG dynamics may involve non-linear effects that a linear model cannot fully capture. These inherent distributional differences across datasets necessitate a cautious interpretation of the findings to ensure that the identified EEG features are robustly attributed to disease states. Future studies may incorporate sensitivity analyses using narrower, age-matched cohorts to better characterize the non-linear interactions between aging and disease-specific neurophysiological signatures. Second, the small sample size—limited by the smaller of the two groups—may have led to a Type II error in the statistical analysis. While the current findings remain significant after rigorous FDR correction, the modest sample size suggests that these results should be viewed as preliminary evidence of shared neurophysiological markers between the two disorders. In addition, because a small sample size may affect the stability of the observed effects, these findings should be interpreted with caution until further validation, such as through larger cohorts or resampling-based analyses, is conducted. Finally, due to the inherent constraints of the publicly available databases, certain socio-demographic and clinical variables, including socioeconomic status, educational level, and detailed medication history for all participants, were not provided. The absence of these variables limits the interpretability of the results and should be considered when generalizing these findings to broader clinical populations.

5. Conclusions

Many studies have identified common pathways in the physiological mechanisms underlying PD and MDD. This study identified features common to the two conditions through analysis of resting-state EEG data, with the potential for diagnostic applications in subsequent evaluations. Our findings indicate that, compared with HCs, both patient groups exhibited significant alterations across multiple EEG features. These included changes in theta, alpha and beta relative power, as well as sample entropy in the delta (centroparietal, temporal, and parietal areas), theta (parieto-occipital area), and gamma (central area) bands. Furthermore, interhemispheric asymmetry was observed across all frequency bands, particularly within the frontal and centroparietal regions. Consistent trends observed in both MDD and PD included increased theta power, decreased alpha and beta power, along with increased parieto-occipital and reduced gamma entropy at FCz. These EEG changes are associated with the underlying physiological mechanisms and clinical manifestations of the two diseases. Overall, this study provides preliminary evidence that PD and MDD share specific EEG characteristics, offering insights into potential common physiological indicators for these conditions. Establishing such commonality offers a theoretical foundation for the future development of cross-disease transfer learning frameworks. While these shared features may eventually facilitate the identification of pathological states, it is essential to distinguish these observed statistical associations from clinically actionable findings at this stage. By leveraging this shared foundation, future research may employ transfer learning to streamline model development for various neurological conditions, including rare diseases with limited data. This hierarchical approach, first identifying shared foundations and then refining for differential discrimination, represents a promising future perspective for enhancing early diagnosis and health management, although substantial clinical validation remains necessary.

Author Contributions

Conceptualization, C.-Y.Y. and H.-Y.C.; methodology, C.-Y.Y. and F.-N.K.; software, F.-N.K.; formal analysis, H.-Y.C.; resources, C.-Y.Y.; writing—original draft preparation, F.-N.K.; writing—review and editing, C.-Y.Y.; visualization, F.-N.K.; supervision, H.-Y.C.; funding acquisition, C.-Y.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported in part by the National Science and Technology Council (NSTC 111-2221-E-033-055-MY3), Taiwan.

Data Availability Statement

The data used in this study are openly available in Patient Repository of EEG Data Computational Tools at [http://predict.cs.unm.edu/ accessed on 1 March 2023].

Acknowledgments

The authors would like to thank Yin-Zhen Chen for her valuable assistance with the EEG signal processing.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Abbreviations

The following abbreviations are used in this manuscript:
PDParkinson’s disease
EEGElectroencephalography
HCsHealthy controls
MDDMajor depressive disorder
DSM-5Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition
BDIBeck depression inventory
ROSReactive oxygen species
HPAHypothalamic–pituitary–adrenal
DFADetrended fluctuation analysis
ShEnShannon entropy
CDCorrelation dimension
BPBand power
FAAFrontal alpha asymmetry
GABAGamma-aminobutyric acid

References

  1. World Health Organization. Parkinson Disease. 2023. Available online: https://www.who.int/news-room/fact-sheets/detail/parkinson-disease (accessed on 19 March 2024).
  2. Guedes, B.F.S.; Cardoso, S.M.; Esteves, A.R. The impact of microRNAs on mitochondrial function and immunity: Relevance to Parkinson’s disease. Biomedicines 2023, 11, 1349. [Google Scholar] [CrossRef] [PubMed]
  3. Schapira, A.H.V.; Chaudhuri, K.R.; Jenner, P. Non-motor features of Parkinson disease. Nat. Rev. Neurosci. 2017, 18, 435–450. [Google Scholar] [CrossRef] [PubMed]
  4. Yuvaraj, R.; Murugappan, M.; Omar, M.I.; Ibrahim, N.M.; Sundaraj, K.; Mohamad, K.; Satiyan, M. Emotion processing in Parkinson’s disease: An EEG spectral power study. Int. J. Neurosci. 2014, 124, 491–502. [Google Scholar] [CrossRef]
  5. Iyer, K.K.; Au, T.R.; Angwin, A.J.; Copland, D.A.; Dissanayaka, N.N. Theta and gamma connectivity is linked with affective and cognitive symptoms in Parkinson’s disease. J. Affect. Disord. 2020, 277, 875–884. [Google Scholar] [CrossRef]
  6. Pappalettera, C.; Miraglia, F.; Cotelli, M.; Rossini, P.M.; Vecchio, F. Analysis of complexity in the EEG activity of Parkinson’s disease patients by means of approximate entropy. GeroScience 2022, 44, 1599–1607. [Google Scholar] [CrossRef] [PubMed]
  7. World Health Organization. Depression. 2021. Available online: https://www.who.int/news-room/fact-sheets/detail/depression (accessed on 19 March 2024).
  8. Pitsillou, E.; Bresnehan, S.M.; Kagarakis, E.A.; Wijoyo, S.J.; Liang, J.; Hung, A.; Karagiannis, T.C. The cellular and molecular basis of major depressive disorder: Towards a unified model for understanding clinical depression. Mol. Biol. Rep. 2019, 47, 753–770. [Google Scholar] [CrossRef]
  9. Mumtaz, W.; Xia, L.; Ali, S.S.A.; Yasin, M.A.M.; Hussain, M.; Malik, A.S. Electroencephalogram (EEG)-based computer-aided technique to diagnose major depressive disorder (MDD). Biomed. Signal Process. Control 2017, 31, 108–115. [Google Scholar] [CrossRef]
  10. Liu, S.; Liu, X.; Yan, D.; Chen, S.; Liu, Y.; Hao, X.; Ou, W.; Huang, Z.; Su, F.; He, F.; et al. Alterations in patients with first-episode depression in the eyes-open and eyes-closed conditions: A resting-state EEG study. IEEE Trans. Neural Syst. Rehabil. Eng. 2022, 30, 1019–1029. [Google Scholar] [CrossRef]
  11. Wu, Z.; Zhong, X.; Lin, G.; Peng, Q.; Zhang, M.; Zhou, H.; Wang, Q.; Chen, B.; Ning, Y. Resting-state electroencephalography of neural oscillation and functional connectivity patterns in late-life depression. J. Affect. Disord. 2022, 316, 169–176. [Google Scholar] [CrossRef]
  12. Yang, L.; Youngblood, H.; Wu, C.; Zhang, Q. Mitochondria as a target for neuroprotection: Role of methylene blue and photobiomodulation. Transl. Neurodegener. 2020, 9, 19. [Google Scholar] [CrossRef]
  13. Duarte, P.; Cuadrado, A.; Leon, R. Monoamine oxidase inhibitors: From classic to new clinical approaches. In Handbook of Experimental Pharmacology; Springer: Cham, Switzerland, 2020; Volume 264, pp. 229–259. [Google Scholar] [CrossRef]
  14. Yeung, S.S.H.; Ho, Y.S.; Chang, R.C.C. The role of meningeal populations of type II innate lymphoid cells in modulating neuroinflammation in neurodegenerative diseases. Exp. Mol. Med. 2021, 53, 1251–1267. [Google Scholar] [CrossRef]
  15. Xu, Y.; Li, Y.; Wang, C.; Han, T.; Liu, H.; Sun, L.; Hong, J.; Hashimoto, M.; Wei, J. The reciprocal interactions between microglia and T cells in Parkinson’s disease: A double-edged sword. J. Neuroinflamm. 2023, 20, 33. [Google Scholar] [CrossRef]
  16. Cavanagh, J.F.; Kumar, P.; Mueller, A.A.; Richardson, S.P.; Mueen, A. Diminished EEG habituation to novel events effectively classifies Parkinson’s patients. Clin. Neurophysiol. 2018, 129, 409–418. [Google Scholar] [CrossRef]
  17. Cavanagh, J.F.; Bismark, A.W.; Frank, M.J.; Allen, J.J.B. Multiple dissociations between comorbid depression and anxiety on reward and punishment processing: Evidence from computationally informed EEG. Psychiatry Res. 2019, 3, 1–17. [Google Scholar] [CrossRef] [PubMed]
  18. Yuvaraj, R.; Murugappan, M. Hemispheric asymmetry non-linear analysis of EEG during emotional responses from idiopathic Parkinson’s disease patients. Cogn. Neurodyn. 2016, 10, 225–234. [Google Scholar] [CrossRef]
  19. Yuvaraj, R.; Murugappan, M.; Ibrahim, N.M.; Omar, M.I.; Sundaraj, K.; Mohamad, K.; Palaniappan, R.; Mesquita, E.; Satiyan, M. On the analysis of EEG power, frequency and asymmetry in Parkinson’s disease during emotion processing. Behav. Brain Funct. 2014, 10, 10–12. [Google Scholar] [CrossRef] [PubMed]
  20. Betrouni, N.; Delval, A.; Chaton, L.; Defebvre, L.; Duits, A.; Moonen, A.; Leentjens, A.F.G.; Dujardin, K. Electroencephalography-based machine learning for cognitive profiling in Parkinson’s disease: Preliminary results. Mov. Disord. 2019, 34, 210–217. [Google Scholar] [CrossRef]
  21. Pezard, L.; Jech, R.; Ruzicka, E. Investigation of non-linear properties of multichannel EEG in the early stages of Parkinson’s disease. Clin. Neurophysiol. 2001, 112, 38–45. [Google Scholar] [CrossRef]
  22. Obukhov, Y.V.; Gabova, A.V.; Zaljalova, Z.A.; Illarioshkin, S.N.; Karabanov, A.V.; Korolev, M.S.; Kuznetsova, G.D.; Morozov, A.A.; Nigmatullina, R.R.; Obukhov, K.Y.; et al. Electroencephalograms features of the early stage Parkinson’s disease. Pattern Recognit. Image Anal. 2014, 24, 593–604. [Google Scholar] [CrossRef]
  23. Chaturvedi, M.; Hatz, F.; Gschwandtner, U.; Bogaarts, J.G.; Meyer, A.; Fuhr, P.; Roth, V. Quantitative EEG (QEEG) measures differentiate Parkinson’s disease (PD) patients from healthy controls (HC). Front. Aging Neurosci. 2017, 9, 3. [Google Scholar] [CrossRef]
  24. Saikia, A.; Hussain, M.; Barua, A.R.; Paul, S. EEG-EMG correlation for Parkinson’s disease. Int. J. Eng. Adv. Technol. 2019, 8, 1179–1185. [Google Scholar] [CrossRef]
  25. Prabhu, K.S.; Martis, R.J. Diagnosis of Parkinson’s disease using computer-aided tool based on EEG. In 2020 IEEE: 17th India Council International Conference (INDICON), New Delhi, India, 10–13 December 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1–4. [Google Scholar] [CrossRef]
  26. Aljalal, M.; Aldosari, S.A.; AlSharabi, K.; Abdurraqeeb, A.M.; Alturki, F.A. Parkinson’s disease detection from resting-state EEG signals using common spatial pattern, entropy, and machine learning techniques. Diagnostics 2022, 12, 1033. [Google Scholar] [CrossRef]
  27. Averna, A.; Coelli, S.; Ferrara, R.; Cerutti, S.; Priori, A.; Bianchi, A.M. Entropy and fractal analysis of brain-related neurophysiological signals in Alzheimer’s and Parkinson’s disease. J. Neural Eng. 2023, 20, 051001. [Google Scholar] [CrossRef]
  28. Yang, C.Y.; Chen, H.Y. Analysis and Discrimination of electroencephalography features for Parkinson’s disease during the induction of negative mood. J. Med. Biol. Eng. 2023, 43, 386–393. [Google Scholar] [CrossRef]
  29. Mahato, S.; Goyal, N.; Ram, D.; Paul, S. Detection of depression and scaling of severity using six-channel EEG data. J. Med. Syst. 2020, 44, 2. [Google Scholar] [CrossRef] [PubMed]
  30. Saeedi, M.; Saeedi, A.; Maghsoudi, A. Major depressive disorder assessment via enhanced k-nearest neighbor method and EEG signals. Phys. Eng. Sci. Med. 2020, 43, 1007–1018. [Google Scholar] [CrossRef]
  31. Nassibi, A.; Papavassiliou, C.; Atashzar, S.F. Depression diagnosis using machine intelligence based on spatiospectrotemporal analysis of multi-channel EEG. Med. Biol. Eng. Comput. 2022, 60, 3187–3202. [Google Scholar] [CrossRef]
  32. de Aguiar Neto, F.S.; Rosa, J.L.G. Depression biomarkers using non-invasive EEG: A review. Neurosci. Biobehav. Rev. 2019, 105, 83–93. [Google Scholar] [CrossRef] [PubMed]
  33. Khosla, A.; Khandnora, P.; Chand, T. Automated diagnosis of depression from EEG signals using traditional and deep learning approaches: A comparative analysis. Biocybern. Biomed. Eng. 2022, 42, 108–142. [Google Scholar] [CrossRef]
  34. Halje, P.; Brys, I.; Mariman, J.J.; Cunha, C.D.; Fuentes, R.; Petersson, P. Oscillations in cortico-basal ganglia circuits: Implications for Parkinson’s disease and other neurologic and psychiatric conditions. J. Neurophysiol. 2019, 122, 203–231. [Google Scholar] [CrossRef] [PubMed]
  35. Bockova, M.; Rektor, I. Impairment of brain functions in Parkinson’s disease reflected by alterations in neural connectivity in EEG studies: A viewpoint. Clin. Neurophysiol. 2019, 130, 239–247. [Google Scholar] [CrossRef]
  36. Grin-Yatsenko, V.A.; Baas, I.; Ponomarev, V.A.; Kropotov, J.D. EEG power spectra at early stages of depressive disorders. J. Clin. Neurophysiol. 2009, 26, 401–406. [Google Scholar] [CrossRef] [PubMed]
  37. Özçoban, M.A.; Tan, O. Electroencephalographic markers in major depressive disorder: Insights from absolute, relative power, and asymmetry analyses. Front. Psychiatry 2025, 15, 1480228. [Google Scholar] [CrossRef]
  38. Roh, S.C.; Kim, J.S.; Kim, S.; Kim, Y.; Lee, S.H. Frontal alpha asymmetry moderated by suicidal ideation in patients with major depressive disorder: A comparison with healthy individuals. Clin. Psychopharmacol. Neurosci. 2020, 18, 58–66. [Google Scholar] [CrossRef]
  39. da Rocha, T.F.; Costa, V.; Camargo, L.; Fernandes, E.B.; Gianlorenço, A.C. Resting-state EEG alpha asymmetry as a potential marker of clinical features in Parkinson’s disease. J. Pers. Med. 2025, 15, 291. [Google Scholar] [CrossRef]
  40. Marcu, G.M.; Szekely-Copîndean, R.D.; Radu, A.M.; Bucuță, M.D.; Fleacă, R.S.; Tănăsescu, C.; Roman, M.D.; Boicean, A.; Băcilă, C.I. Resting-state frontal, frontlateral, and parietal alpha asymmetry: A pilot study examining relations with depressive disorder type and severity. Front. Psychol. 2023, 14, 1087081. [Google Scholar] [CrossRef]
  41. Luo, Y.; Tang, M.; Fan, X. Meta-analysis of resting frontal alpha asymmetry as a biomarker of depression. npj Ment. Health Res. 2025, 4, 2. [Google Scholar] [CrossRef] [PubMed]
  42. Lau, Z.J.; Pham, T.; Chen, S.H.A.; Makowski, D. Brain entropy, fractal dimensions and predictability: A review of complexity measures for EEG in healthy and neuropsychiatric populations. Eur. J. Neurosci. 2022, 56, 5047–5069. [Google Scholar] [CrossRef]
  43. Faust, O.; Ang, P.C.A.; Puthankattil, S.D.; Joseph, P.K. Depression diagnosis support system based on EEG signal entropies. J. Mech. Med. Biol. 2014, 14, 1450035. [Google Scholar] [CrossRef]
  44. Kanunikov, I.E.; Kleeva, D.F. Electroencephalography entropy and depression. S.S. Korsakov J. Neurol. Psychiatry 2022, 122, 106–110. [Google Scholar] [CrossRef]
  45. Yi, G.S.; Wang, J.; Deng, B.; Wei, X.L. Complexity of resting-state EEG activity in the patients with early-stage Parkinson’s disease. Cogn. Neurodyn. 2017, 11, 147–160. [Google Scholar] [CrossRef] [PubMed]
  46. Keller, S.M.; Gschwandtner, U.; Meyer, A.; Chaturvedi, M.; Roth, V.; Fuhr, P. Cognitive decline in Parkinson’s disease is associated with reduced complexity of EEG at baseline. Brain Commun. 2020, 2, fcaa207. [Google Scholar] [CrossRef] [PubMed]
  47. Rahmati, M.; Jalaeianbanayan, A.; Vahedi, J.; Najafzadeh, H.; Danesh, B. Novel EEG-based diagnostic framework for Major Depressive Disorder using microstate and entropy features. Cogn. Neurodyn. 2025, 19, 116. [Google Scholar] [CrossRef]
  48. Lee, C.H.; Juan, C.H.; Chen, H.H.; Hong, J.P.; Liao, T.W.; French, I.; Lo, Y.S.; Wang, Y.R.; Cheng, M.L.; Wu, H.C.; et al. Long-Range Temporal Correlations in Electroencephalography for Parkinson’s Disease Progression. Mov. Disord. 2025, 40, 266–275. [Google Scholar] [CrossRef]
  49. Salomon, R.M.; Cowan, R.L. Oscillatory serotonin function in depression. Synapse 2013, 67, 801–820. [Google Scholar] [CrossRef]
  50. Nyström, C.; Matousek, M.; Hällström, T. Relationships between EEG and biochemical parameters in major depressive disorder. Acta Psychiatr. Scand. 1988, 77, 457–462. [Google Scholar] [CrossRef]
  51. Olbrich, S.; Trankner, A.; Chittka, T.; Hegerl, U.; Schonknecht, P. Functional connectivity in major depression: Increased phase synchronization between frontal cortical EEG-source estimates. Psychiatry Res. Neuroimaging 2014, 222, 91–99. [Google Scholar] [CrossRef] [PubMed]
  52. Ip, C.T.; Bardeci, M.D.; Kronenberg, G.; Pinborg, L.H.; Seifritz, E.; Brunovsky, M.; Olbrich, S. EEG-vigilance regulation is associated with and predicts ketamine response in major depressive disorder. Transl. Psychiatry 2024, 14, 64. [Google Scholar] [CrossRef]
  53. Buddhala, C.; Loftin, S.K.; Kuley, B.M.; Cairns, N.J.; Campbell, M.C.; Perlmutter, J.S.; Kotzbauer, P.T. Dopaminergic, serotonergic, and noradrenergic deficits in Parkinson disease. Ann. Clin. Transl. Neurol. 2015, 2, 949–959. [Google Scholar] [CrossRef]
  54. Kemp, A.F.; Kinnerup, M.; Johnsen, B.; Jakobsen, S.; Nahimi, A.; Gjedde, A. EEG frequency correlates with α2-receptor density in Parkinson’s disease. Biomolecules 2024, 14, 209. [Google Scholar] [CrossRef]
  55. Schreiner, S.J.; Imbach, L.L.; Valko, P.O.; Maric, A.; Maqkaj, R.; Werth, E.; Baumann, C.R.; Baumann-Vogel, H. Reduced regional NREM sleep slow-wave activity is associated with cognitive impairment in Parkinson disease. Front. Neurol. 2021, 12, 618101. [Google Scholar] [CrossRef] [PubMed]
  56. Waninger, S.; Berka, C.; Karic, M.S.; Korszen, S.; Mozley, P.D.; Henchcliffe, C.; Kang, Y.; Hesterman, J.; Mangoubi, T.; Verma, A. Neurophysiological biomarkers of Parkinson’s disease. J. Park. Dis. 2020, 10, 471–480. [Google Scholar] [CrossRef]
Signals 07 00046 g001
Figure 1. Overall flowchart of feature analysis.
Figure 1. Overall flowchart of feature analysis.
Signals 07 00046 g001
Signals 07 00046 g005
Figure 5. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of DFA between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD.
Figure 5. Topographic maps of EEG power differences with electrode location profiles, based on statistical analysis of DFA between patients (PD or MDD) and HCs. White circles indicate electrodes significant for PD or MDD.
Signals 07 00046 g005
Signals 07 00046 g006
Figure 6. Significant hemispheric asymmetry (right power−left power) between electrode pairs in patients with PD or MDD and HCs across delta (a), theta (b), alpha (c), beta (d), and gamma (e) bands. Gray labels indicate electrode pairs with no significant differences in either PD or MDD. Black labels indicate electrode pairs with significant differences for PD or MDD, while red labels indicate electrode pairs with significant differences for both conditions.
Figure 6. Significant hemispheric asymmetry (right power−left power) between electrode pairs in patients with PD or MDD and HCs across delta (a), theta (b), alpha (c), beta (d), and gamma (e) bands. Gray labels indicate electrode pairs with no significant differences in either PD or MDD. Black labels indicate electrode pairs with significant differences for PD or MDD, while red labels indicate electrode pairs with significant differences for both conditions.
Signals 07 00046 g006
Table 1. Demographic characteristics (mean and standard deviation) of the patient and healthy controls.
Table 1. Demographic characteristics (mean and standard deviation) of the patient and healthy controls.
DatasetGroupGenderAgeBDIUPDRS
PD datasetPD13 M: 7 F69.75 (8.59)7.64 (5.23)22.25 (7.92)
HC11 M: 9 F69.30 (9.40)5.27 (4.20)-
MDD datasetMDD8 M: 12 F18.91 (1.34)21.52 (5.66)-
HC7 M: 13 F18.82 (1.96)1.14 (1.36)-
Table 2. Characteristic items frequently analyzed in the previous literature.
Table 2. Characteristic items frequently analyzed in the previous literature.
DiseaseSelected FeaturesReference
PDAbsolute power density, ApEn, Band power (BP), Correlation dimension (CD), DFA, Fractal dimension, Higher order spectrum, Hurst exponent, Interhemispheric (absolute) power asymmetry, Kolmogorov complexity, Largest Lyapunov exponent, Log energy entropy, Localized entropy, Lyapunov exponent, Mean frequency, Multiscale entropy, Norm entropy, Permutation entropy, Relative power, Relative power density, Relative spectral power, SampEn, Shannon entropy (ShEn), Slope asymmetry, Sure entropy and Threshold entropy[18,19,20,21,22,23,24,25,26,27,28]
MDDAlpha asymmetry, Alpha power variability, ApEn, BP, Brain laterality, C0-complexity, CD, Cluster coefficient, Coherence, DFA, Functional connectivity, Higuchi’s fractal dimension, Hjorth activity, Hjorth complexity, Hjorth mobility, Interhemispheric asymmetry, Kolmogorov-entropy, Lempel–Ziv complexity, Maximum Lyapunov exponent, Mean frequency, Median frequency, Paired asymmetry, Path-length, Phase lag index, Phase synchronization, Power-spectrum density, Power-spectral entropy, Rayleigh-entropy, Relative band power, Root-mean-square, SampEn, ShEn, Singular-value deposition entropy and Spectral asymmetry index[29,30,31,32,33]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yang, C.-Y.; Kuo, F.-N.; Chen, H.-Y. Exploratory Analysis of Electroencephalography Characteristics Shared by Major Depressive Disorder and Parkinson’s Disease: A Database Study. Signals 2026, 7, 46. https://doi.org/10.3390/signals7030046

AMA Style

Yang C-Y, Kuo F-N, Chen H-Y. Exploratory Analysis of Electroencephalography Characteristics Shared by Major Depressive Disorder and Parkinson’s Disease: A Database Study. Signals. 2026; 7(3):46. https://doi.org/10.3390/signals7030046

Chicago/Turabian Style

Yang, Chia-Yen, Fan-Ning Kuo, and Hsin-Yung Chen. 2026. "Exploratory Analysis of Electroencephalography Characteristics Shared by Major Depressive Disorder and Parkinson’s Disease: A Database Study" Signals 7, no. 3: 46. https://doi.org/10.3390/signals7030046

APA Style

Yang, C.-Y., Kuo, F.-N., & Chen, H.-Y. (2026). Exploratory Analysis of Electroencephalography Characteristics Shared by Major Depressive Disorder and Parkinson’s Disease: A Database Study. Signals, 7(3), 46. https://doi.org/10.3390/signals7030046

Article Metrics

Back to TopTop