Daria Migotina

Symbolic representation of the EEG for sleep stage classification

Manual visualization-based sleep stage classification is a time-consuming task prone to errors. Since the correct identification of sleep stages is vital for the correct identification of sleep disorders and for the research in this field in general, there is a growing demand for efficient automatic classification methods. However, there is still no symbolic representation of the biomedical signals that leads to a reliable and accurate automatic sleep classification system. This work presents the application of a novel method for symbolic representation of the EEG and evaluates its potential as information source for a sleep stage classifier, in this case a SVM classifier. The data is first analyzed using Self-Organizing Maps (SOM) and a mutual information (MI)-based variable selection algorithm. Preliminary results of sleep data classification provide success rates around 70%. These results are promising since only EEG is used, and there is still room for improvement in this new symbolic representation of the signal.

EEG training platform: Improving Brain-Computer Interaction and cognitive skills

This work proposes a complete structure of an EEG biofeedback platform focused on an efficient way for its user to learn how to self regulate cortical activity. A longitudinal study of how voluntary training of specific electro cortical activity produces any stable changes in the electroencephalogram is also presented. Correlations of these changes with short term memory are also hypothesized. In this work the human brain was seen as an electrochemical machine capable of receiving stimuli and adapt accordingly. So, only relevant EEG activity was fed back to the trainee by a Brain Computer Interface (BCI) in an intelligible way, allowing the identification of phasic changes in the EEG and what cognitive state caused it. The results from this study showed that it is possible to learn to change some rhythmical activity in the EEG, in this case the alpha activity, after a few feedback sessions. A positive relation between this frequency band and cognitive processes was also observed. These results also indicate that the proposed EEG biofeedback protocol can be shaped as a tool for those who rely on BCI to communicate with the external world as it allows the tracking and training of individual frequency bands.

Automatic Artifacts Detection and Classification in Sleep EEG Signals Using Descriptive Statistics and Histogram Analysis: Comparison of Two Detectors

The algorithm for artifacts detection and classification, applying different sets of constraint rules, was proposed. Two automatic artifacts detectors based on the proposed algorithm and thresholding techniques that use descriptive statistics and a histogram analysis, were developed. At first, the performance of both detectors was evaluated by matching with the human expert scoring. Detectors were tested with various threshold values; and ones that provided the best performance results were selected. At the end, two detectors were compared with each other and the best detector was identified. Detected artifacts were classified into three different types: body movement, muscle and slow eye movement artifacts. Analyses of detection and classification results were performed separately for the whole night sleep and for NREM sleep stages 1, 2, 3 and 4. The artifacts detector, based on the application of a threshold, calculated from a histogram, has provided the best detection results with approximately 85% of sensitivity and 84% of correct classification; the best classification results were obtained for body movement artifacts, with approximately 94% of specificity and 99% of correct classification in the analysis, performed for NREM sleep stages 1, 2, 3, and 4.

Object recognition test in peripheral vision: a study on the influence of object color, pattern and shape

As an important factor for central vision preview, peripheral vision is a crucial ability for most ball game players in motion detection. A critical problem with peripheral vision is object recognition which has not yet been given much attention. This paper presents an experimental study to evaluate the influence on object recognition in peripheral vision due to different patterns, colors and shapes of the objects. More specifically, four types of shapes (including circles, triangles, horizontal stripes and vertical stripes) with various colors presented in different patterns were applied during the peripheral vision test. The results show that different patterns and colors indeed affect object recognition in peripheral vision in terms of accuracy and response time, while different types of shapes do not vary the performance significantly.

Automatic K-complex detection using Hjorth parameters and fuzzy decision

K-complex is a stereotyped transient wave in the human electroencephalogram (EEG), it appears frequently during sleep recordings. Its role and significance have been disregarded since its discovery until recently, when the American Association of Sleep Medicine (AASM) proposed a new classification of sleep with a relevant role for the K-complexes in the definition of the sleep stages. It is now one of the key features that contribute to sleep stage assessment. K-complexes are associated with sleep arousal and can occur spontaneously or as an evoked response to external stimuli. Since the EEG has a stochastic nature, K-complex can have a wide variety of shapes and it can be difficult to distinguish it from other EEG waves. The visual scoring of K-complex is a very complex, time consuming and expensive procedure and reported agreement between human expert scorers is very poor. In this paper a new method for K-complex automatic detection is presented. The detection method is based on Hjorth parameters and employs fuzzy decision. The performance of the detection system is compared to the visual human scoring.

Dynamic peripheral visual performance relates to alpha activity in soccer players

Many studies have demonstrated the relationship between the alpha activity and the central visual ability, in which the visual ability is usually assessed through static stimuli. Besides static circumstance, however in the real environment there are often dynamic changes and the peripheral visual ability in a dynamic environment (i.e., dynamic peripheral visual ability) is important for all people. So far, no work has reported whether there is a relationship between the dynamic peripheral visual ability and the alpha activity. Thus, the objective of this study was to investigate their relationship. Sixty-two soccer players performed a newly designed peripheral vision task in which the visual stimuli were dynamic, while their EEG signals were recorded from Cz, O1, and O2 locations. The relationship between the dynamic peripheral visual performance and the alpha activity was examined by the percentage-bend correlation test. The results indicated no significant correlation between the dynamic peripheral visual performance and the alpha amplitudes in the eyes-open and eyes-closed resting condition. However, it was not the case for the alpha activity during the peripheral vision task: the dynamic peripheral visual performance showed significant positive inter-individual correlations with the amplitudes in the alpha band (8–12 Hz) and the individual alpha band (IAB) during the peripheral vision task. A potential application of this finding is to improve the dynamic peripheral visual performance by up-regulating alpha activity using neuromodulation techniques.

Sonification of Sleep EEG

This paper introduces a new methodology for converting sleep Electroencephalogram (EEG) signals into sound. The main goal is to investigate the possibility of encoding sleep events into sequences of notes and breaks, generating musical sound that is consistent and audible, while allowing a global appraisal of sleep dynamics.