A5040565095- EEG and Brain-Computer Interfaces
- Neural dynamics and brain function
- Neuroscience and Neural Engineering
- Functional Brain Connectivity Studies
- Blind Source Separation Techniques
- Bayesian Modeling and Causal Inference
- Neural and Behavioral Psychology Studies
- Neural Networks and Applications
- Muscle activation and electromyography studies
- Advanced Memory and Neural Computing
- Explainable Artificial Intelligence (XAI)
- Gaze Tracking and Assistive Technology
- Motor Control and Adaptation
- Stroke Rehabilitation and Recovery
- Machine Learning and Data Classification
- Neurological disorders and treatments
- Advanced Graph Neural Networks
- Transcranial Magnetic Stimulation Studies
- Biomedical Text Mining and Ontologies
- Bioinformatics and Genomic Networks
- Cognitive Science and Mapping
- Cell Image Analysis Techniques
- Error Correcting Code Techniques
- Machine Learning and Algorithms
- Statistical Methods and Inference
University of Vienna
2019-2024
Complexity Science Hub Vienna
2021-2024
Statistics Austria
2023
Science Hub
2021-2023
Max Planck Institute for Intelligent Systems
2011-2021
Max Planck Society
2009-2018
Ludwig-Maximilians-Universität München
2017-2018
Sabancı Üniversitesi
2016
University of Zurich
2013
ETH Zurich
2013
The performance of brain-computer interfaces (BCIs) improves with the amount available training data, statistical distribution this however, varies across subjects as well sessions within individual subjects, limiting transferability data or trained models between them. In article, we review current transfer learning techniques in BCIs that exploit shared structure multiple and/or to increase performance. We then present a framework for context can be applied any arbitrary feature space,...
We address two shortcomings of the common spatial patterns (CSP) algorithm for filtering in context brain--computer interfaces (BCIs) based on electroencephalography/magnetoencephalography (EEG/MEG): First, question optimality CSP terms minimal achievable classification error remains unsolved. Second, has been initially proposed two-class paradigms. Extensions to multiclass paradigms have suggested, but are heuristics. these framework information theoretic feature extraction (ITFE). show...
Analyzing neural signals and providing feedback in realtime is one of the core characteristics a brain-computer interface (BCI). As this feature may be employed to induce plasticity, utilizing BCI technology for therapeutic purposes increasingly gaining popularity community. In paper, we discuss state-of-the-art research on topic, address principles challenges inducing plasticity by means BCI, delineate problems study design outcome evaluation arising context. We conclude with list open...
Many methods for causal inference generate directed acyclic graphs (DAGs) that formalize relations between $n$ variables. Given the joint distribution on all these variables, DAG contains information about how intervening one variable changes of other $n-1$ However, quantifying influence another remains a nontrivial question. Here we propose set natural, intuitive postulates measure strength should satisfy. We then introduce communication scenario, where edges in play role channels can be...
The combination of brain–computer interfaces (BCIs) with robot-assisted physical therapy constitutes a promising approach to neurorehabilitation patients severe hemiparetic syndromes caused by cerebrovascular brain damage (e.g. stroke) and other neurological conditions. In such scenario, key aspect is how reestablish the disrupted sensorimotor feedback loop. However, date it an open question artificially closing loop influences decoding performance BCI. this paper, we answer issue studying...
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Spatial filtering (SF) constitutes an integral part of building EEG-based brain–computer interfaces (BCIs). Algorithms frequently used for SF, such as common spatial patterns (CSPs) and independent component analysis, require labeled training data identifying filters that provide information on a subject's intention, which renders these algorithms susceptible to overfitting artifactual EEG...
The brain exhibits a complex temporal structure which translates into hierarchy of distinct neural timescales. An open question is how these intrinsic timescales are related to sensory or motor information processing and whether dynamics have common patterns in different behavioral states. We address questions by investigating the brain's healthy controls, (amyotrophic lateral sclerosis, locked-in syndrome), (anesthesia, unresponsive wakefulness progressive reduction (from awake states over...
Motor-imagery-based brain-computer interfaces (BCIs) commonly use the common spatial pattern filter (CSP) as preprocessing step before feature extraction and classification. The CSP method is a supervised algorithm therefore needs subject-specific training data for calibration, which very time consuming to collect. In order reduce amount of calibration that needed new subject, one can apply multitask (from now on called multisubject) machine learning techniques phase. Here, goal multisubject...
Even though feedback is considered to play an important role in learning how operate a brain-computer interface (BCI), date no significant influence of design on BCI-performance has been reported literature. In this work, we adapt standard motor-imagery BCI-paradigm study affected by biasing the belief subjects have their level control over BCI system. Our findings indicate that already capable operating are impeded inaccurate feedback, while normally performing or close chance may actually...
Subjects operating a brain–computer interface (BCI) based on sensorimotor rhythms exhibit large variations in performance over the course of an experimental session. Here, we show that high-frequency γ-oscillations, originating fronto-parietal networks, predict such trial-to-trial basis. We interpret this finding as empirical support for influence attentional networks BCI via modulation rhythm.
Interpretable Machine Learning (IML) methods are used to gain insight into the relevance of a feature interest for performance model. Commonly IML differ in whether they consider features isolation, e.g., Permutation Feature Importance (PFI), or relation all remaining variables, Conditional (CFI). As such, perturbation mechanisms inherent PFI and CFI represent extreme reference points. We introduce Relative (RFI), generalization that allows more nuanced importance computation beyond versus...
A neurorehabilitation approach that combines robot-assisted active physical therapy and Brain-Computer Interfaces (BCIs) may provide an additional mileage with respect to traditional rehabilitation methods for patients severe motor impairment due cerebrovascular brain damage (e.g., stroke) other neurological conditions. In this paper, we describe the design modes of operation a robot-based framework enables artificial support sensorimotor feedback loop. The aim is increase cortical...
Abstract Objective Inhibitory control has been discussed as a developmental and maintenance factor in binge‐eating disorder (BED). The current study is the first aimed at investigating inhibitory negative mood condition on psychophysiological behavioral level BED with combination of electroencephalography (EEG) eye tracking (ET). Method We conducted combined EEG ET overweight individuals (BED+, n = 24, mean age 31, BMI 35 kg/m 2 ) without (BED–, 23, 28, normal‐weight (NWC, 26, 22 group....
Brain-computer interface (BCI) systems are often based on motor- and/or sensory processes that known to be impaired in late stages of amyotrophic lateral sclerosis (ALS). We propose a novel BCI designed for patients ALS only requires high-level cognitive transmit information from the user BCI.We trained subjects via EEG-based neurofeedback self-regulate amplitude gamma-oscillations superior parietal cortex (SPC). argue likely associated with attentional processes, thereby providing...
Objective. Recent brain-computer interface (BCI) assisted stroke rehabilitation protocols tend to focus on sensorimotor activity of the brain. Relying evidence claiming that a variety brain rhythms beyond areas are related extent motor deficits, we propose identify neural correlates learning spatially and spectrally for further use in novel BCI-assisted neurorehabilitation settings. Approach. Electroencephalographic (EEG) data were recorded from healthy subjects participating physical...