A self-organizing maps classifier structure for brain computer interfaces

Bueno, Lívia Maronesi; Bastos, Teodiano

doi:10.1590/2446-4740.0753

Cited by 7 publications

(4 citation statements)

References 21 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SOMs have also been used to provide interpretable models of the respiratory signals by considering the neurons in the map as identifiers of specific internal states of the dynamical system generating the respiratory timeseries [12]. A similar approach has been taken by [8,24] where they have been used to identify and visualize the different mental states from engineered features extracted from Electroencephalography data. SOMs have also been used in stress detection tasks [14] where they are fed with features from skin conductance and ECG data and the resulting maps are clustered by a Gaussian Mixture Model to identify SOM units that can be associated with relax phases or stress phases, in a fully unsupervised way.…”

Section: Self Organizing Maps In Biosignal Processingmentioning

confidence: 99%

Topographic mapping for quality inspection and intelligent filtering of smart-bracelet data

Bacciu

Bertoncini

Morelli

2021

Neural Comput & Applic

View full text Add to dashboard Cite

Wrist-worn wearable devices equipped with heart activity sensors can provide valuable data that can be used for preventative health. However, hearth activity analysis from these devices suffers from noise introduced by motion artifacts. Methods traditionally used to remove outliers based on motion data can yield to discarding clean data, if some movement was present, and accepting noisy data, i.e. subject was still but the sensor was misplaced. This work shows that Self Organizing Maps (SOMs) can be used to effectively accept or reject sections of heart data collected from unreliable devices, such as wrist worn devices. In particular, the proposed SOM-based filter can accept a larger amount of measurements (less false negatives) with an higher overall quality with respect to methods solely based on statistical analysis of motion data. We provide an empirical analysis on real world wearable data, comprising heart and motion data of users. We show how topographic mapping can help identifying and interpreting patterns in the sensor data and help relating them to an assessment of user state. More importantly, our experimental results shows the proposed approach is able to retain almost twice the amount of data while keeping samples with an error that is an order of magnitude lower with respect to a filter based on accelerometric data.

show abstract

Section: Self Organizing Maps In Biosignal Processingmentioning

confidence: 99%

Topographic mapping for quality inspection and intelligent filtering of smart-bracelet data

Bacciu

Bertoncini

Morelli

2021

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…As a classification method, SOM was used to identify epileptic condition [10], emotions [11] and other mental states [12,13,14]. For brain-computer interfaces, SOMs are used to classify motor imagery [15,16,17] and other mental tasks [18]. Being also a dimensionality reduction technique, SOM allows to visualize the recorded neural signal for the benefit of the the user or a researcher.…”

Section: Related Workmentioning

confidence: 99%

Mental state space visualization for interactive modeling of personalized BCI control strategies

et al. 2020

View full text Add to dashboard Cite

Objective. Numerous studies in the area of BCI are focused on the search for a better experimental paradigm -a set of mental actions that a user can evoke consistently and a machine can discriminate reliably. Examples of such mental activities are motor imagery, mental computations, etc. We propose a technique that instead allows the user to try different mental actions in the search for the ones that will work best. Approach. The system is based on a modification of the self-organizing map (SOM) algorithm and enables interactive communication between the user and the learning system through a visualization of user's mental state space. During the interaction with the system the user converges on the paradigm that is most efficient and intuitive for that particular user. Main results. Results of the two experiments, one allowing muscular activity, another permitting mental activity only, demonstrate soundness of the proposed method and offer preliminary validation of the performance improvement over the traditional closed-loop feedback approach. Significance. The proposed method allows a user to visually explore their mental state space in real time, opening new opportunities for scientific inquiry. The application of this method to the area of brain-computer interfaces enables more efficient search for the mental states that will allow a user to reliably control a BCI system.

show abstract

“…The EEG classification is the main procedure when implementing a BCI process, and the expected results depend greatly on it [8,[13][14][15][16]. It makes it possible to classify the user's wishes using a training procedure depending on data characteristics [6,14]. Linear discriminant analysis (LDA) [17], which is commonly used for the BCI system, is suitable for real time implementation and requires only a few computational resources.…”

Section: Introductionmentioning

confidence: 99%

“…The second level, called beta rhythm, has a frequency of 14-30 Hz and appears under conditions of active awakening and sleep. The third level, which is the theta rhythm (4-7 Hz), appears during the installation of a sleep phase, followed by the delta rhythm (0.5-3 Hz), which is characteristic of sleeping adults [4][5][6].…”

Section: Introductionmentioning

confidence: 99%

Neural Net-Based Approach to EEG Signal Acquisition and Classification in BCI Applications

et al. 2019

View full text Add to dashboard Cite

The following contribution describes a neural net-based, noninvasive methodology for electroencephalographic (EEG) signal classification. The application concerns a brain–computer interface (BCI) allowing disabled people to interact with their environment using only brain activity. It consists of classifying user’s thoughts in order to translate them into commands, such as controlling wheelchairs, cursor movement, or spelling. The proposed method follows a functional model, as is the case for any BCI, and can be achieved through three main phases: data acquisition and preprocessing, feature extraction, and classification of brains activities. For this purpose, we propose an interpretation model implementing a quantization method using both fast Fourier transform with root mean square error for feature extraction and a self-organizing-map-based neural network to generate classifiers, allowing better interpretation of brain activities. In order to show the effectiveness of the proposed methodology, an experimental study was conducted by exploiting five mental activities acquired by a G.tec BCI system containing 16 simultaneously sampled bio-signal channels with 24 bits, with experiments performed on 10 randomly chosen subjects.

show abstract

A self-organizing maps classifier structure for brain computer interfaces

Cited by 7 publications

References 21 publications

Topographic mapping for quality inspection and intelligent filtering of smart-bracelet data

Topographic mapping for quality inspection and intelligent filtering of smart-bracelet data

Mental state space visualization for interactive modeling of personalized BCI control strategies

Neural Net-Based Approach to EEG Signal Acquisition and Classification in BCI Applications

Contact Info

Product

Resources

About