1. Assessment of model accuracy in eyes open and closed EEG data : effect of data pre-processing and validation methodsMattiev Jamolbek Maqsudovich, Jakob Sajovic, Gorazd Drevenšek, Peter Rogelj, 2023, izvirni znanstveni članek Povzetek: Eyes open and eyes closed data is often used to validate novel human brain activity classification methods. The cross-validation of models trained on minimally preprocessed data is frequently utilized, regardless of electroencephalography data comprised of data resulting from muscle activity and environmental noise, affecting classification accuracy. Moreover, electroencephalography data of a single subject is often divided into smaller parts, due to limited availability of large datasets. The most frequently used method for model validation is cross-validation, even though the results may be affected by overfitting to the specifics of brain activity of limited subjects. To test the effects of preprocessing and classifier validation on classification accuracy, we tested fourteen classification algorithms implemented in WEKA and MATLAB, tested on comprehensively and simply preprocessed electroencephalography data. Hold-out and cross-validation were used to compare the classification accuracy of eyes open and closed data. The data of 50 subjects, with four minutes of data with eyes closed and open each was used. The algorithms trained on simply preprocessed data were superior to the ones trained on comprehensively preprocessed data in cross-validation testing. The reverse was true when hold-out accuracy was examined. Significant increases in hold-out accuracy were observed if the data of different subjects was not strictly separated between the test and training datasets, showing the presence of overfitting. The results show that comprehensive data preprocessing can be advantageous for subject invariant classification, while higher subject-specific accuracy can be attained with simple preprocessing. Researchers should thus state the final intended use of their classifier.
Ključne besede: electroencephalography (EEG), machine learning, model validation
Objavljeno v DiRROS: 04.08.2026; Ogledov: 114; Prenosov: 64
 Celotno besedilo (5,68 MB)
Gradivo ima več datotek! Več... |
2. Melatonin vs. dexmedetomidine for sleep induction in children before electroencephalographyKatja Peganc, David Neubauer, Jasna Oražem, Mirjana Perković-Benedik, Urška Mahne, Neli Bizjak, Zvonka Rener-Primec, Nataša Šuštar, Tita Butenko, Eva Vrščaj, Damjan Osredkar, 2024, izvirni znanstveni članek Ključne besede: melatonin, dexmedetomidine, electroencephalography, EEG, sleep, sleep induction
Objavljeno v DiRROS: 08.06.2026; Ogledov: 250; Prenosov: 90
Celotno besedilo (625,42 KB) |
3. The complexity of caffeine’s effects on regular coffee consumersMateja Lesar, Jakob Sajovic, Dušanka Novaković, Maša Primožič, Eva Vetrih, Martin Sajovic, Anja Žnidaršič, Peter Rogelj, Andreas Daffertshofer, Zoran Levnajić, Gorazd Drevenšek, 2025, izvirni znanstveni članek Povzetek: Why does coffee wake us up? Is it because it contains caffeine, or because we are used to it waking us up after drinking it? To answer this question, we recruited twenty habitual coffee drinkers who received either caffeinated or decaffeinated coffee (placebo) in a double-blind, randomized fashion. The two substances were identical except for the presence of caffeine. We measured cognitive performance, cardiovascular responses, and whole-head EEG during rest and during an auditory-oddball task. The same measurements were done before and after ingestion. We expected to find significant differences between caffeine and placebo groups across the outcome measures. However, except for the resting-state alpha power, changes due to ingestion in physiological responses and in cognitive functioning were not significantly different between the two groups. Actually, only one of the three cognitive measures was found to be significantly altered by the ingestion. These findings suggest that regular coffee consumers respond to coffee-like beverages independently of the presence of caffeine.
Ključne besede: caffeine, decaffeine, ERP, event related potentials (ERP), electroencephalography (EEG), EEG, auditory odball, cognitive performance, resting state EEG
Objavljeno v DiRROS: 14.04.2026; Ogledov: 337; Prenosov: 250
Celotno besedilo (2,45 MB)
Gradivo ima več datotek! Več... |
4. The interplay of brain neurotransmission and mental fatigue : a research protocolY. L. Arenales Arauz, Jelle Habay, Tjaša Ocvirk, Ana Mali, Suzanna Russell, Uroš Marušič, Kevin De Pauw, Bart Roelands, 2024, drugi znanstveni članki Povzetek: Introduction Mental fatigue (MF) significantly affects both cognitive and physical performance. However, the precise mechanisms, particularly concerning neurotransmission, require further investigation. An implication of the role of dopamine (DA) and noradrenaline (NA) is stated, but empirical evidence for this theory still needs to be provided. To address this gap, we aim to investigate the role of brain neurotransmission in elucidating if, and how prolonged cognitive activity induces MF and its subsequent impact on cognitive performance. Methods This study (registration number: G095422N) will adopt a randomized cross-over design with sixteen healthy participants aged 18–35 years. The sessions include a familiarization, two experimental (DA: 20mg Methylphenidate; NA: 8mg Reboxetine) conditions, and one placebo (lactose tablet: 10mg) condition. A 60-minute individualized Stroop task will be used to investigate whether, and how the onset of MF changes under the influence of reuptake inhibitors. Attention and response inhibition will be assessed before and after the MF-inducing task using a Go/NoGo task. The integration of physiological (electroencephalography, heart rate), behavioral (attention, response inhibition), and subjective indicators (scales and questionnaires) will be used to detect the underlying mechanisms holistically. Data analysis will involve linear mixed models with significance at p<0.05. Discussion The integration of diverse techniques and analyses offers a comprehensive perspective on the onset and impact of MF, introducing a novel approach. Future research plans involve extending this protocol to explore the connection between brain neurotransmission and physical fatigue. This protocol will further advance our understanding of the complex interplay between the brain and fatigue.
Ključne besede: mental fatigue, neurotransmission, cognitive performance, dopamine, noradrenaline, electroencephalography (EEG)
Objavljeno v DiRROS: 10.02.2026; Ogledov: 719; Prenosov: 274
Celotno besedilo (1,30 MB)
Gradivo ima več datotek! Več... |
5. |
6. Exploring the impact of electroencephalography-based neurofeedback (EEG NFB) on motor deficits in Parkinson’s disease : a targeted literature reviewLaura Blaznik, Uroš Marušič, 2025, pregledni znanstveni članek Povzetek: Parkinson’s disease (PD) is a progressive neurodegenerative disorder, with pharmacological treatments predominantly focusing on dopaminergic therapies. In the early stages of PD, symptoms may also be alleviated through non-pharmacological interventions. One such non-invasive technique is electroencephalogram neurofeedback (EEG NFB), which has shown promising results in improving the cognitive and motor functions of PD patients. The aim of our study was to assess the existing evidence, identify key trends and determine potential opportunities for future research in the field of EEG NFB for PD. This analysis explores the impact of EEG NFB on motor deficits in PD and identifies key factors for the successful implementation of EEG NFB as evidenced in the literature. The synthesis includes findings from five relevant studies, including one case study, one pilot study and three randomized controlled trials. Study selection followed the PICO framework to ensure relevance and rigor. The results suggest a correlation between sensorimotor rhythm (SMR) and beta rhythms, with increases in SMR (13–15 Hz) and beta (12–15 Hz) rhythms linked to improvements in balance, mobility and stability in PD patients. However, limitations such as small sample sizes, brief intervention durations and lack of follow-up warrant a cautious interpretation. Future research should prioritize robust protocols, larger samples and extended neurofeedback training to fully assess EEG NFB’s potential for PD management.
Ključne besede: Parkinson's disease, motor deficits, biofeedback, neurofeedback, electroencephalography
Objavljeno v DiRROS: 27.02.2025; Ogledov: 1156; Prenosov: 682
Celotno besedilo (371,13 KB)
Gradivo ima več datotek! Več... |
7. Cortico-muscular phase connectivity during an isometric knee extension task in people with early Parkinson’s diseaseNina Omejc, Tomislav Stankovski, Manca Peskar, Miloš Kalc, Paolo Manganotti, Klaus Gramann, Sašo Džeroski, Uroš Marušič, 2025, izvirni znanstveni članek Povzetek: — Introduction: Parkinson’s disease (PD) is characterized by enhanced beta-band activity (13–30 Hz) in the motor control regions. Simultaneously, corticomuscular (CM) connectivity in the beta-band during isometric contractions tends to decline with age, in various diseases, and under dual-task conditions. Objective: This study aimed to characterize electroencephalograph (EEG) and electromyograph (EMG) power spectra during a motor task, assess CM phase connectivity, and explore how these measures are modulated by an additional cognitive task. Specifically, we focused on the beta-band to explore the relationship between heightened beta amplitude and reduced beta CM connectivity. Methodology: Early-stage people with PD and age-matched controls performed an isometric knee extension task, a cognitive task, and a combined dual task, while EEG (128ch) and EMG (2x32ch) were recorded. CM phase connectivity was assessed through phase coherence and a phase dynamics model. Results: The EEG power spectrum revealed no cohort differences in the beta-band. EMG also showed no differences up to 80 Hz. However, the combined EEG-EMG analysis uncovered reduced beta phase coherence in people with early PD during the motor task. CM phase coherence exhibited distinct scalp topography and frequency ranges compared to the EEG power spectrum, suggesting different mechanisms for pathological beta increase and CM connectivity. Additionally, phase dynamics modelling indicated stronger directional coupling from the cortex to the active muscle and less prominent phase coupling across people with PD. Despite high inter-individual variability, these metrics may prove useful for personalized assessments, particularly in people with heightened CM connectivity.
Ključne besede: electroencephalography, brain modeling, electromiography, coherence, motors, diseases, couplings, electrodes, oscillators, protocols
Objavljeno v DiRROS: 13.01.2025; Ogledov: 1270; Prenosov: 883
Celotno besedilo (3,42 MB)
Gradivo ima več datotek! Več... |
8. Beta–gamma phase-amplitude coupling as a non-invasive biomarker for Parkinson’s diseas : insights from Electroencephalography studiesTisa Hodnik, Stiven Roytman, Nicolaas I. Bohnen, Uroš Marušič, 2024, pregledni znanstveni članek Povzetek: Phase-amplitude coupling (PAC) describes the interaction of two separate frequencies in which the lower frequency phase acts as a carrier frequency of the higher frequency amplitude. It is a means of carrying integrated streams of information between micro- and macroscale systems in the brain, allowing for coordinated activity of separate brain regions. A beta–gamma PAC increase over the sensorimotor cortex has been observed consistently in people with Parkinson’s disease (PD). Its cause is attributed to neural entrainment in the basal ganglia, caused by pathological degeneration characteristic of PD. Disruptions in this phenomenon in PD patients have been observed in the resting state as well as during movement recordings and have reliably distinguished patients from healthy participants. The changes can be detected non-invasively with the electroencephalogram (EEG). They correspond to the severity of the motor symptoms and the medication status of people with PD. Furthermore, a medication-induced decrease in PAC in PD correlates with the alleviation of motor symptoms measured by assessment scales. A beta–gamma PAC increase has, therefore, been explored as a possible means of quantifying motor pathology in PD. The application of this parameter to closed-loop deep brain stimulation could serve as a self-adaptation measure of such treatment, responding to fluctuations of motor symptom severity in PD. Furthermore, phase-dependent stimulation provides a new precise method for modulating PAC increases in the cortex. This review offers a comprehensive synthesis of the current EEG-based evidence on PAC fluctuations in PD, explores the potential practical utility of this biomarker, and provides recommendations for future research.
Ključne besede: neurodegenerative diseases, Parkinson’s disease, electroencephalography, phase-amplitude coupling
Objavljeno v DiRROS: 21.03.2024; Ogledov: 1587; Prenosov: 1042
Celotno besedilo (604,09 KB)
Gradivo ima več datotek! Več... |
9. A data-driven machine learning approach for brain-computer interfaces targeting lower limb neuroprostheticsArnau Dillen, Elke Lathouwers, Aleksandar Miladinović, Uroš Marušič, Fakhreddine Ghaffari, Olivier Romain, Romain Meeusen, Kevin De Pauw, 2022, izvirni znanstveni članek Povzetek: Prosthetic devices that replace a lost limb have become increasingly performant in recent years. Recent advances in both software and hardware allow for the decoding of electroencephalogram (EEG) signals to improve the control of active prostheses with brain-computer interfaces (BCI). Most BCI research is focused on the upper body. Although BCI research for the lower extremities has increased in recent years, there are still gaps in our knowledge of the neural patterns associated with lower limb movement. Therefore, the main objective of this study is to show the feasibility of decoding lower limb movements from EEG data recordings. The second aim is to investigate whether well-known neuroplastic adaptations in individuals with an amputation have an influence on decoding performance. To address this, we collected data from multiple individuals with lower limb amputation and a matched able-bodied control group. Using these data, we trained and evaluated common BCI methods that have already been proven effective for upper limb BCI. With an average test decoding accuracy of 84% for both groups, our results show that it is possible to discriminate different lower extremity movements using EEG data with good accuracy. There are no significant differences (p = 0.99) in the decoding performance of these movements between healthy subjects and subjects with lower extremity amputation. These results show the feasibility of using BCI for lower limb prosthesis control and indicate that decoding performance is not influenced by neuroplasticity-induced differences between the two groups.
Ključne besede: neuroprosthetics, brain-computer interface, machine learning, electroencephalography, data-driven learning, lower limb amputation
Objavljeno v DiRROS: 21.07.2022; Ogledov: 2195; Prenosov: 1366
Celotno besedilo (858,15 KB)
Gradivo ima več datotek! Več... |
10. Higher neural demands on stimulus processing after prolonged hospitalization can be mitigated by a cognitively stimulating environmentUroš Marušič, Rado Pišot, Voyko Kavcic, 2021, izvirni znanstveni članek Povzetek: Prolonge d periods of complete physical inactivity or bed rest trigger various alterations in the functional and metabolic levels of the human body. However, bed rest-related adaptations of the central nervous system are less known and thoroughly studied. The aim of this study was to investigate brain electrophysiological changes using event-related potentials (ERPs) after 14 days of bed rest and 12 consecutive sessions of computerized cognitive training (CCT). Sixteen older (Mage= 60 years) healthy volunteers were randomly divided into a CCT treatment group and an active control group. All participants performed ERP measurements based on the foveal visual presentation of a circle on a black background before and after bed rest. After 14 days of bed rest, participants in the control group showed increased peak P1 amplitude (p = .012), decreased P1 latency (p = .024), and increased P2 amplitude (p = .036), while the CCT group also showed decreased P1 latency (p = .023) and decreased P2 latency (p = .049). Our results suggest that, even from a central adaptation perspective, prolonged periods of physical inactivity or bed rest trigger additional neural recruitment and should therefore be minimized, and that CCT may serve as a tool to mitigate this. Future research should focus on other aspects of central nervous system adaptation following periods of immobilization/hospitalization to improve our knowledge of infl uence of physical inactivity and its eff ects on cortical activity and to develop appropriate countermeasures to mitigate functional dysregulation.
Ključne besede: aging, physical inactivity, immobilization, electroencephalography, EEG, computerized cognitive training
Objavljeno v DiRROS: 13.05.2021; Ogledov: 2600; Prenosov: 2056
Celotno besedilo (820,21 KB)
Gradivo ima več datotek! Več... |
