Artificial neural network simulations of event-related potentials generation using massively parallel computer machines

1994 ◽  
Vol 18 (2) ◽  
pp. 156
Author(s):  
Porto Carras
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Event Related Potentials ◽  
Parallel Computer ◽  
Massively Parallel ◽  
Related Potentials ◽  
Network Simulations ◽  
Artificial Neural

scholarly journals Modeling Nonlinear Transfer Functions from Speech Envelopes to Encephalography with Neural Networks

2019 ◽  
Vol 11 (4) ◽  
pp. 1 ◽  
Author(s):  
Tobias de Taillez ◽  
Florian Denk ◽  
Bojana Mirkovic ◽  
Birger Kollmeier ◽  
Bernd T. Meyer
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Linear Models ◽  
Transfer Functions ◽  
Event Related Potentials ◽  
Stimulus Presentation ◽  
Related Potentials ◽  
Linear And Nonlinear ◽  
Artificial Neural

Diferent linear models have been proposed to establish a link between an auditory stimulus and the neurophysiological response obtained through electroencephalography (EEG). We investigate if non-linear mappings can be modeled with deep neural networks trained on continuous speech envelopes and EEG data obtained in an auditory attention two-speaker scenario. An artificial neural network was trained to predict the EEG response related to the attended and unattended speech envelopes. After training, the properties of the DNN-based model are analyzed by measuring the transfer function between input envelopes and predicted EEG signals by using click-like stimuli and frequency sweeps as input patterns. Using sweep responses allows to separate the linear and nonlinear response components also with respect to attention. The responses from the model trained on normal speech resemble event-related potentials despite the fact that the DNN was not trained to reproduce such patterns. These responses are modulated by attention, since we obtain significantly lower amplitudes at latencies of 110 ms, 170 ms and 300 ms after stimulus presentation for unattended processing in contrast to the attended. The comparison of linear and nonlinear components indicates that the largest contribution arises from linear processing (75%), while the remaining 25% are attributed to nonlinear processes in the model. Further, a spectral analysis showed a stronger 5 Hz component in modeled EEG for attended in contrast to unattended predictions. The results indicate that the artificial neural network produces responses consistent with recent findings and presents a new approach for quantifying the model properties.

scholarly journals Simulation Models of the Seawater Greenhouse

2018 ◽  
Vol 7 (3.16) ◽  
pp. 90
Author(s):  
Abdulrahim M. Al-Ismaili ◽  
Tahani Bait Suwailam
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
High Temperatures ◽  
Water Scarcity ◽  
Agricultural Development ◽  
Simulation Models ◽  
Factors Affecting ◽  
Network Simulations ◽  
Artificial Neural ◽  
Seawater Greenhouse

In arid climates, extremely high temperatures in the summer and the chronic water scarcity put a firm barrier against agricultural development and sustainability. The SWGH technology is an engineering phenomenon that came to overcome both the constraints particularly in areas where seawater is accessible and/or brackish groundwater is available. It is a greenhouse used to cultivate crops and at the same time produce its own freshwater need. This study aimed to highlight the models that were carried out to simulate the SWGH as a whole or only the dehumidification rate of the SWGH condenser. Four types of simulation models were identified, namely, analytical, numerical, empirical and artificial neural network simulations. The factors affecting the dehumidification rate were also discussed taking into consideration the results from the simulation models.  

scholarly journals Detection of Error-Related Potentials in Stroke Patients from EEG Using an Artificial Neural Network

Sensors ◽  
2021 ◽  
Vol 21 (18) ◽  
pp. 6274
Author(s):  
Nayab Usama ◽  
Imran Khan Niazi ◽  
Kim Dremstrup ◽  
Mads Jochumsen
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Stroke Rehabilitation ◽  
Retest Reliability ◽  
Linear Discriminant ◽  
Temporal Features ◽  
Continuous Eeg ◽  
Related Potentials ◽  
Artificial Neural ◽  
Test Retest Reliability

Error-related potentials (ErrPs) have been proposed as a means for improving brain–computer interface (BCI) performance by either correcting an incorrect action performed by the BCI or label data for continuous adaptation of the BCI to improve the performance. The latter approach could be relevant within stroke rehabilitation where BCI calibration time could be minimized by using a generalized classifier that is continuously being individualized throughout the rehabilitation session. This may be achieved if data are correctly labelled. Therefore, the aims of this study were: (1) classify single-trial ErrPs produced by individuals with stroke, (2) investigate test–retest reliability, and (3) compare different classifier calibration schemes with different classification methods (artificial neural network, ANN, and linear discriminant analysis, LDA) with waveform features as input for meaningful physiological interpretability. Twenty-five individuals with stroke operated a sham BCI on two separate days where they attempted to perform a movement after which they received feedback (error/correct) while continuous EEG was recorded. The EEG was divided into epochs: ErrPs and NonErrPs. The epochs were classified with a multi-layer perceptron ANN based on temporal features or the entire epoch. Additionally, the features were classified with shrinkage LDA. The features were waveforms of the ErrPs and NonErrPs from the sensorimotor cortex to improve the explainability and interpretation of the output of the classifiers. Three calibration schemes were tested: within-day, between-day, and across-participant. Using within-day calibration, 90% of the data were correctly classified with the entire epoch as input to the ANN; it decreased to 86% and 69% when using temporal features as input to ANN and LDA, respectively. There was poor test–retest reliability between the two days, and the other calibration schemes led to accuracies in the range of 63–72% with LDA performing the best. There was no association between the individuals’ impairment level and classification accuracies. The results show that ErrPs can be classified in individuals with stroke, but that user- and session-specific calibration is needed for optimal ErrP decoding with this approach. The use of ErrP/NonErrP waveform features makes it possible to have a physiological meaningful interpretation of the output of the classifiers. The results may have implications for labelling data continuously in BCIs for stroke rehabilitation and thus potentially improve the BCI performance.

scholarly journals Optimasi Parameter Input pada Artificial Neural Network untuk Meningkatkan Akurasi Prediksi Indeks Harga Saham

2021 ◽  
Vol 10 (2) ◽  
pp. 211-216
Author(s):  
Ignatius Wiseto Prasetyo Agung
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Stock Price ◽  
Technology Development ◽  
Moving Average ◽  
Stock Index ◽  
Accurate Analysis ◽  
Input Parameters ◽  
Network Simulations ◽  
Artificial Neural

Stock trading is one of the businesses that has been done worldwide. In order to gain the maximum profit, accurate analysis is needed, so a trader can decide to buy and sell stock at the perfect time and price. Conventionally, two analyses are employed, namely fundamental and technical.  Technical analysis is obtained based on historical data that is processed mathematically. Along with technology development, stock price analysis and prediction can be performed with the help of computational algorithms, such as machine learning. In this research, Artificial Neural Network simulations to produce accurate stock price predictions were carried out. Experiments are performed by using various input parameters, such as moving average filters, in order to produce the best accuracy. Simulations are completed with stock index datasets that represent three continents, i.e. NYA (America, USA), GDAXI (Europe, Germany), and JKSE (Asia, Indonesia). This work proposes a new method, which is the utilization of input parameters combinations of C, O, L, H, MA-5 of C, MA-5 of O, and the average of O C prices. Furthermore, this proposed scheme is also compared to previous work done by Khorram et al, where this new work shows more accurate results.

Sign in / Sign up

Export Citation Format

Share Document