Showing papers by "Gert Pfurtscheller published in 1993"

Alpha frequency, cognitive load and memory performance.

Abstract: EEG-signals were recorded from subjects as they performed a modified version of Schneider's and Shiffrin's memory search paradigm. The hypothesis was tested whether individual (centre of gravity) alpha frequency, termed IAF, is related to memory performance and/or attentional demands. The results show that memory performance exerts the strongest effect on IAF. As compared to a resting period, the difference in IAF between age-matched good and bad memory performers reached a maximum when subjects were actually retrieving information from their memory. During retrieval, the IAF of good performers is 1.25 Hz higher than for bad performers. Attentional and task demands also tend to reduce IAF, but as compared to memory performance-to a much lesser degree. The results of amplitude analyses demonstrate further that during retrieval, alpha desynchronization is more pronounced for bad performers than for good performers. Taken together, the results indicate that a decrease in IAF is always related to a drop in performance.

Discovering Patterns in EEG-Signals: Comparative Study of a Few Methods

Abstract: The objective of this paper is to draw the attention of the ML-researchers to the domain of data analysis. The issue is illustrated by an attractive case study—automatic classification of non-averaged EEG-signals. We applied several approaches and obtained best results from a combination of an ID3-like program with Bayesian learning.

Graz brain-computer interface: an EEG-based cursor control system

Abstract: .Abstmcf ~ The idea of n new cnnimunicntion device a BninComputer Interface (BCI) for handicnppd penons is io record bioelectrical signals (EEC) on the intact scalp and use the cln%sificaiion of these signals for control or devices. This pnper discusses ongoing research in this neld and reports on a system developed in C m where EEC. i s used to control n cumor on n ntoniior in one dimension.

Computer-based analysis of continuous non-invasive blood pressure and heart rate variability-methodology and normal values during wakefulness and sleep.

Abstract: Polysomnographic recordings were made in 10 healthy male adult subjects (mean age 25.1 +/- 2.8 years). Parameters were obtained from continuous non-invasive measurement of blood pressure and from heart rate spectra analysed both in the waking state and during sleep (stage 4). The total heart rate variability (mean +/- SE) was not significantly diminished (p < 0.2; paired t-test) during stage 4 sleep (6.2 +/- 0.5%) as compared with the waking state (7.1 +/- 0.5%). The relative heart rate variability coefficient within the frequency band HRV-II ((3-9)/min) was, however, significantly higher (p < 0.001) during slow-wave sleep (0.55 +/- 0.04) than during wakefulness (0.34 +/- 0.04). This fact is in accordance with results concerning an estimated value of the baroreflex sensitivity, which was also significantly higher (p < 0.001) during sleep stage 4 (7.0 +/- 0.8 ms/mmHg vs. 10.4 +/- 1.4 mmHg).

Computer controlled brain death documentation in the intensive care unit

Abstract: An interactive, knowledge-based computer system for brain death documentation is presented. The specific exponents BRAINDEX R and G were realised by the software tool Personal Consultant Plus and the programming language Clipper, respectively. The strategies of conclusion were forward chaining for approximate evaluation of coma stages and backward chaining for analysing the brain death syndrome. BRAINDEX was developed for use with an IBM personal computer or compatible equipment. Systemic analyses were compared retrospectively with the data from clinical brain death protocols (n = 132) of 128 comatose patients (mean age 35.1 +/- 15.8 years) with a Glasgow Coma Score of 3. Identical classifications (system vs physician) were found in all patients without diagnosis of brain death (n = 35). Differences related to the findings of the physician were evaluated in lower numbers of the systemic positive diagnosis of brain death (82 vs 89) and higher numbers of impossibility of systemic evaluation (11 vs 2). These results were obtained by conclusions of the computer system drawn by restrictive systemic mechanisms to avoid false-negative diagnoses. The system therefore seems to be useful for documentation, consultation, and as a teaching instrument and data bank in brain death.

On-line Learning With Learning Vector Quantization; A Case Study Of EEG Classification

Abstract: Real-time classification, e.g. of EEG, is one possible application of the Learning Vector Quantizer (LVQ) [1]. Its main advantage over other classifiers is its simplicity and speed but also the possibility for on-line learning. Usually, real-time EEG classifiers must be created off-line on the basis of a seperate recording. It would be preferable if the classifier could create itself on-line in a training session, i.e. start from a very sub-optimal initial state, e.g. using a classifier of another subject, and train itself on-line. Data recorded in three subjects during sessions, where a cursor was controlled in real-time based on EEG classification (Graz Brain-Computer Interface, BCI) [2], are examined in off-line simulations. Each subject participated in 4–6 sessions and the subject-dependent LVQs were updated between sessions to improve their generalization ability. Training LVQs on these data sets with varying values of the learning parameter α, suitable parameter ranges are derived for performances which are comparable to those obtained during the recording. The simulation results show that LVQ1 is a learning algorithm which is well suited for on-line learning because of a number of facilities: (1) LVQ is a very fest and well-understood classification method. Speed is an important factor when EEG classification is carried out in real-time. (2) On-line learning can be incorporated without much additional expense: only one reference vector (the winner) must be updated (either drawn further towards the current input vector or pushed slightly away), therefore the additional calculation time only depends on the input vector dimension (in our case: 10) but not on the number of reference vectors used (usuali’ about 8). Note that this would be different if we used a Multi-Layer Perceptron: the error has to be backpropagated through the whole network and therefore the time for calculation depends on the topology (the size) of the network. (3) The amount of update can be controlled via the learning parameter α. As a rule of thumb, α should range between 0.001 and 0.01, depending on the performance in former sessions. For the first session, a big a is preferable to bring the LVQ, which can stem from some other subject, into the right general position of the current user. For the following sessions, α can be lowered to freeze the LVQ in its position. Additional improvement can be expected if both learning during data processing and learning between sessions are combined. Although this study was based only on off-line examination of existing recordings, subjects in future experiments can also be expected to give improved performance: if even the first session gives more than random results they will experience the system as more trustworthy and will be more motivated in the following sessions.

Towards automated sleep classification in infants using symbolic and subsymbolic approaches.

Abstract: The paper addresses the problem of automatic sleep classification. A special effort is made to find a method of extracting reasonable descriptions of the individual sleep stages from sample measurements of EGG, EMG, EOG, etc., and from a classification of these measurements provided by an expert. The method should satisfy three requirements: classification accuracy, interpretability of the results, and the ability to select the relevant and discard the irrelevant variables. The solution suggested in this paper consists of a combination of the subsymbolic algorithm LVQ with the symbolic decision tree generator ID3. Results demonstrating the feasibility and utility of our approach are also presented.

LVQ-Based On-line EEG Classification

Abstract: In this paper the problems of classifying EEG in both its spatial as well as its temporal aspects are described. The task was to identify the intended side of hand movement (left or right) on the basis of EEG recorded on two channels during one second before movement onset. This is part of a larger project aiming to build a “Brain-Computer Interface” (BCI) which should enable handicapped persons to communicate with their surroundings using their EEG.

Diagnostische Möglichkeiten des elektrophysiologischen Monitorings an der Intensivstation

Abstract: EINLEITUNG: Moderne Intensivstationen erlauben es, die lebenswichtigen Funktionen des Herz-KreislaufSystems und der Atmung kontinuierlich zu überwachen. Die Homöostase der Körperflüssigkeiten kann durch kurzfristige Laborkontrollen gewährleistet werden. Die vor allem bei Patienten mit schwerem Schädelhirntrauma wichtige Kontrolle der Funktion des Zentralnervensystems entzog sich jedoch bislang einer kontinuierlichen Überwachung. Nur indirekt konnte durch in kurzen Intervallen durchgeführte klinischneurologische Untersuchungen mit dem Erstellen von Komaskalen oder durch die Messung des intrakraniellen Druckes auf die Integrität der zerebralen Funktionen geschlossen werden. Neurologische Überprüfungen können jedoch, selbst wenn sie engmaschig durchgeführt werden, stets nur punktuellen Charakter haben. Die Messung des intrakraniellen Druckes wiederum stellt einen invasiven Eingriff dar, sodaß die Indikation dazu nur in akuten, offensichtlich lebensbedrohlichen Fällen gestellt wird.

Chaos oder Rauschen in EEG Signalen - Eine chaostheoretische Untersuchung

Abstract: I. Einleitung Zur Zeit gibt es bezüglich der Eigenschaften des EEG Signales zwei verschiedene Betrachtungsweisen. Entweder wird das EEG als zufälliges Rauschen interpretiert, oder man betrachtet es als Ausgangssignal eines nichtlinearen dynamischen Systems, das chaotisches Verhalten auftveist und durch einen deterministisch chaotischen Attraktor beschrieben werden kann [1]. Die Dynamik eines solchen Systems bzw. seine Freiheitsgrade können durch das Korrelationsintegral eines Signales dieses Systems beschrieben werden. Aus der Steigung des Korrelationsintegrales kann dann die Korrelationsdimension geschätzt werden [2,3]. Ein großes Problem bei der Bestimmung der Korrelationsdimension von EEG Signalen stellt allerdings das S/N Verhältnis der Signale dar. Daher muß besonderer Wert auf die Datenvorverarbeitung gelegt werden. Ein wichtiger Punkt dabei ist die Verbesserung des S/N Verhältnisses. Meistens sind lineare Filter ausreichend, um die Qualität der Daten zu verbessern. Aber speziell bei der Analyse mit chaostheoretischen Methoden (Bestimmung der Korrelationsdimension) kann durch lineare Filterung wichtige Information verlorengehen [4,5J. Um dieses Problem zu umgehen, wurde eine spezielle, nichtlineare Filtermethode entwickelt. Dieser Filteralgorithmus basiert auf einer nicht stationären Zustandsraummodellierung und nützt dabei die numerischen Vorteile der sogenannten Hauptkomponentenmethode mit Hilfe der Singulärwertezerlegung [5.61. Im speziellen wurden EEG Signale untersucht, bei denen ausgeprägte Rhythmen im 10 Hz Bereich (a und ) auftreten. Dabei ergaben sich nach der Filterung folgende Ergebnisse: Die Korrelationsdimension der und Rhythmen ist niedrig (zwischen 5 und 7). was durchaus auf chaotisches Verhalten hindeuten kann.

Koma-Prognostik mittels SSEP- und AEHP-Monitoring

Abstract: Die Befunde der AEHP waren gekennzeichnet durch zahlreiche Komponentenverluste bzw. Latenzverlängerungen. Das Amplitudenverhältnis Va/Ia war prognostisch gesehen aussagekräftiger als die I-V Interpeaklatenz (Abb. 4). EINLEITUNG: Prognostische Beurteilungen von Patienten mit schweren Hirnschäden sind aufgrund der Komplexität interferierender Einflüsse oft schwer zu treffen. Elektrophysiologische Untersuchungsverfahren liefern dabei eine wertvolle Unterstützung. Ziel dieser Studie ist, die prognostische Anwendbarkeit von frühen Komponenten somatosensorisch evozierter Potentiale (SSEP) und akustisch evozierter Hirnstammpotentiale (AEHP) am komatösen Intensivpatienten zu prüfen.