Automatic recognition of alertness and drowsiness from EEG by an artificial neural network (original) (raw)
Related papers
Artificial neural network for detecting drowsiness from EEG recordings
6th Seminar on Neural Network Applications in Electrical Engineering, 2002
We present a novel method for classifying alert vs drowsy states from 1 s long sequences of full spectrum EEG recordings in an arbitrary subject. This novel method uses time series of interhemispheric and intrahemispheric cross spectral densities of full spectrum EEG as the input to an artificial neural network (ANN) with two discrete outputs: drowsy and alert. The experimental data were collected from 17 subjects. Two experts in EEG interpretation visually inspected the data and provided the necessary expertise for the training of an ANN. We selected the following three ANNs as potential candidates: (1) the linear network with Widrow-Hoff (WH) algorithm; (2) the non-linear ANN with the Levenberg-Marquardt (LM) rule; and (3) the Learning Vector Quantization (LVQ) neural network. We showed that the LVQ neural network gives the best classification compared with the linear network that uses WH algorithm (the worst), and the non-linear network trained with the LM rule. Classification properties of LVQ were validated using the data recorded in 12 healthy volunteer subjects, yet whose EEG recordings have not been used for the training of the ANN. The statistics were used as a measure of potential applicability of the LVQ: the t-distribution showed that matching between the human assessment and the network output was 94.37 ± 1.95%. This result suggests that the automatic recognition algorithm is applicable for distinguishing between alert and drowsy state in recordings that have not been used for the training.
We present a novel method for classifying alert vs drowsy states from 1 s long sequences of full spectrum EEG recordings in an arbitrary subject. This novel method uses time series of interhemispheric and intrahemispheric cross spectral densities of full spectrum EEG as the input to an artificial neural network (ANN) with two discrete outputs: drowsy and alert. The experimental data were collected from 17 subjects. Two experts in EEG interpretation visually inspected the data and provided the necessary expertise for the training of an ANN. We selected the following three ANNs as potential candidates: (1) the linear network with Widrow-Hoff (WH) algorithm; (2) the non-linear ANN with the Levenberg-Marquardt (LM) rule; and (3) the Learning Vector Quantization (LVQ) neural network. We showed that the LVQ neural network gives the best classification compared with the linear network that uses WH algorithm (the worst), and the non-linear network trained with the LM rule. Classification properties of LVQ were validated using the data recorded in 12 healthy volunteer subjects, yet whose EEG recordings have not been used for the training of the ANN. The statistics were used as a measure of potential applicability of the LVQ: the t-distribution showed that matching between the human assessment and the network output was 94.37 ± 1.95%. This result suggests that the automatic recognition algorithm is applicable for distinguishing between alert and drowsy state in recordings that have not been used for the training.
IJERT-Neural Network Based Drowsiness Detection Using Electroencephalogram
International Journal of Engineering Research and Technology (IJERT), 2013
https://www.ijert.org/neural-network-based-drowsiness-detection-using-electroencephalogram https://www.ijert.org/research/neural-network-based-drowsiness-detection-using-electroencephalogram-IJERTV2IS100680.pdf Driver drowsiness is one of the main factors in many traffic accidents. It can be detected by monitoring physiological signals in order to detect drowsiness with the change in the patterns of the EOG, EEG and ECG signals. The main issue in such a technique is to extract a set of features that can highly differentiate between the different drowsiness levels. In this work, a new system for driver's drowsiness detection based on EEG using Neural network is proposed. This uses physiological data of drivers to detect drowsiness. These include the measurement of brain wave by using 25 channels EEG and approaches based on EEG signals. EEG data is converted into excel sheet from where we detect the alpha waves, which are indicators of drowsiness. The result shows the Roc graphs and confusion matrix for the samples which gives the combined accuracy result which is 81.8%.
Automatic classification of Sleep Stages on a EEG signal by Artificial Neural Networks
2005
Visual analysis of the physiological signals recorded at sleep time constitutes a heavy task for the clinician. In fact data quantity to be analyzed, generally corresponding to eight hours of recordings studied per 30s epochs, as well as the complexity of this analysis require a significant time. The objective of our work is to propose a tool for automatic analysis and decision-making based on artificial neural networks (ANN). In this paper, we present an outline of this tool and we propose to compare human and ANN performances on a simple case of vigilance states labeling. The first difficulty consists of the choice of representation for the physiological signals and in particular the electroencephalogram (EEG) which is regarded as the principal indicator of sleep stages. Once the representation is adopted, the following step is the design of the optimal ANN by a training and validation process on data set of a healthy adult. The results obtained, on average 76% of agreement betwee...
Automatic recognition of alertness level from EEG by using neural network and wavelet coefficients
Expert Systems with Applications, 2005
Electrophysiological recordings are considered a reliable method of assessing a person's alertness. Sleep medicine is asked to offer objective methods to measure daytime alertness, tiredness and sleepiness. As EEG signals are non-stationary, the conventional method of frequency analysis is not highly successful in recognition of alertness level. This paper deals with a novel method of analysis of EEG signals using wavelet transform, and classification using ANN. EEG signals were decomposed into the frequency sub-bands using wavelet transform and a set of statistical features was extracted from the sub-bands to represent the distribution of wavelet coefficients. Then these statistical features were used as an input to an ANN with three discrete outputs: alert, drowsy and sleep. The error back-propagation neural network is selected as a classifier to discriminate the alertness level of a subject. EEG signals were obtained from 30 healthy subjects. The group consisted of 14 females and 16 males with ages ranging from 18 to 65 years and a mean age of 33.5 years, and a Body Mass Index (BMI) of 32.4G7.3 kg/m 2 . Alertness level and classification properties of ANN were tested using the data recorded in 12 healthy subjects, whereby the EEG recordings were not used to train the ANN. The statistics were used as a measure of potential applicability of the ANN. The accuracy of the ANN was 95G3% alert, 93G4% drowsy and 92G5% sleep. q
Drowsiness detection by analysis of EEG signal with the help of Machine Learning
Conference: 24th annual International Conference on Advanced Computing and Communications (ADCOM 2018)At: International Institute of Information Technology Bangalore(IIITB), 2018
Electroencephalogram (EEG) signals give physiological indications of mind whether you are resting or prepared to do some work, think or dream. To understand these things it is essential to aggregate some EEG motion from the human cerebrum. Among all open physiological indications of the cerebrum, EEG is economical and adaptive. The aim of this work is detecting drowsiness of a person by analyzing EEG signal with the help of Machine learning. For this, we examining the EEG signal and we extract the features. Here three different types of features such as time, frequency-based features and sub-band based features are extracted. Total we extract 66 features from each epoch of individual subject in order to classify drowsiness and alertness. Here we considered total 30 persons and we used the Random forest as the classifier techniques to predict the drowsiness states. With these all parameters, with different ratios of active and drowsy training samples our experimental results show more than 95% accuracy.
Journal of Medical Systems, 2014
There is a wealth of analysis techniques that can be used in analyzing data of such a nature as EEG (Electroencephalogram), yet there are still many more ways and possibilities of analysis techniques to consider in order to produce a method that far exceeds the capabilities of the prevalent method. Since a multilayer neural network with multi-valued neurons (MLMVN) was successfully used earlier to decode EEG signals in a brain/computer interface (BCI) by analysis of their Fourier transform, it seemed very attractive to use it as a tool for EEG analysis. This work aims to further investigate how a complex-valued machine learning tool can be used to analyze EEG in the frequency domain. Our goal was to check how Fourier transform and complex wavelet transform of EEG can be analyzed using MLMVN in order to diagnose epilepsy, its remission or absence. We worked with a commonly used benchmark data set of epilepsy-related EEGs. The analysis of the transformed data was performed to determine a set of relevant statistical characteristics of DTCWT and Fourier transform components, which were then used as inputs of the MLMVN. The obtained results show a very high efficiency of the proposed approach.
Journal of Neuroscience Methods, 2021
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Automatic recognition of alertness level by using wavelet transform and artificial neural network
We propose a novel method for automatic recognition of alertness level from full spectrum electroencephalogram (EEG) recordings. This procedure uses power spectral density (PSD) of discrete wavelet transform (DWT) of full spectrum EEG as an input to an artificial neural network (ANN) with three discrete outputs: alert, drowsy and sleep. The error back propagation neural network is selected as a classifier to discriminate the alertness level of a subject. EEG signals were obtained from 30 healthy subjects. The group consisted of 14 females and 16 males with ages ranging from 18 to 65 years and a mean age of 33.5 years, and a body mass index (BMI) of 32.4 ± 7.3 kg/m 2 . Alertness level and classification properties of ANN were tested using the data recorded in 12 healthy subjects, whereby the EEG recordings were not used been used to train the ANN. The statistics were used as a measure of potential applicability of the ANN. The accuracy of the ANN was 96 ± 3% alert, 95 ± 4% drowsy and 94 ± 5% sleep. The results suggest that the automatic recognition algorithm is applicable for distinguishing between alert, drowsy and sleep state in recordings that have not been used for the training.
Support vector machine based detection of drowsiness using minimum EEG features
Drowsiness presents major safety concerns for tasks that require long periods of focus and alertness. While there is a body of work on drowsiness detection using EEG signals in neuroscience and engineering, there exist unanswered questions pertaining to the best mechanisms to use for detecting drowsiness. Targeting a range of practical safety-awareness applications, this study adopts a machine learning based approach to build support vector machine (SVM) classifiers to distinguish between awake and drowsy states. While broadband alpha, beta, delta, and theta waves are often used as features in the existing work, lack of widely agreed precise definitions of such broadband signals and difficulty in accounting for interpersonal variability has led to poor classification performance as demonstrated in this study. Furthermore, the transition from wakefulness to drowsiness and deeper sleep stages is a complex multifaceted process. The richness of this process calls for inclusion of sub-band features for more accurate drowsiness detection. To shed light on the effectiveness of sub-banding, we quantitatively compare the performances of a large set of SVM classifiers trained upon a varying number of 1Hz subband features. More importantly, we identify a compact set of neuroscientifcally motivated EEG features and demonstrate that the resulting classifier not only outperforms traditional broadband based classifiers but also is on a par with or superior than the best sub-band classifiers found by thorough search in a large space of 1Hz subband features.