A WPCA-Based Method for Detecting Fatigue Driving From EEG-Based Internet of Vehicles System

Brain-Computer Interface (BCI) or Human-Machine Interface now becoming vital biomedical engineering and technology field which applying EEG technologies to provide assistive device technology (AT) to humans. Hence, this paper presents the results of analyzing EEG signals from various human cognitive states to extract the suitable EEG features that can be employed to control BCI devices which can be used by disabled or paralyzed people. The EEG features in term of power spectral density, spectral centroids, standard deviation and entropy are selected and investigated from two different mental exercises; (i) quick solving math and (ii) relax (do nothing). The selected features then are classified using Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) and K-Nearest Neighbors (k-NN) classifier. Among all these features, the best accuracy have been achieved by the power spectral density. The accuracies of this feature are 95%, 100%, 100% with LDA, SVM and K-NN respectively. Finally, the translation algorithm will be constructed using selected and classified EEG features to control the BCI devices.

Driver fatigue is one of the main causes for traffic accidents. It is very important to identify fatigued driving to prevent traffic incidents objectively and accurately. Detecting drivers' physiological signals is an important method of identifying fatigued driving. Moreover, special attention has been paid to physiological signals as methods for detecting fatigues driving, such as with ECG (electroencephalography) and EEG (electrocardiogram) tests. In this paper, ECG and EEG data were collected through experiments conducted through a driving simulator. The ECG trend was obtained based on statistical analysis. Moreover, the authors compared EEG signals under the awake and fatigued states by combining the video and the ECG trend. The EEG characteristics were then extracted, and the EEG threshold value is obtained. The effectiveness of these indices is evaluated in further experiments. The results could be used as the foundation for future studies on driver fatigue.

In this paper, an attempt is made to apply few conventional methods of EEG feature extraction and classification methods and compare their performance for a specific task. Two different feature extraction and classification methods are implemented to classify the mental tasks of EEG signals from a known dataset. For this purpose, the auto regression model and the wavelet transform is used as feature extraction. A combined EEG feature vector is also evaluated on the classification accuracy. The features extracted from these methods are applied to the k-nearest neighbour and support vector machine classifiers separately. Each subject has ten trials of each mental task, in which five trials of each task is used for training the system. The remaining five tasks are used for testing the system. Four different trial combinations of each task are made. The results are evaluated using the confusion matrix. Experimental results specify that each method has specific advantages and disadvantages and is suitable for EEG signal analysis for a specific application.

Fatigue driving is one of the main causes of traffic accidents. If we can predict that a driver will be fatigued in the near future, and give warnings and parking suggestions to let the driver stop and rest, traffic accidents can be reduced effectively. In the past, most research work on fatigue driving used conventional Back Propagation Neural Network (BPNN) model or Support Vector Machine (SVM) to make predictions. BPNN and SVM could not co-relate sequential instances of data in that series, thus making predictions not so accurate. A system is proposed that uses Recurrent Neural Network (RNN) with Long Short-term Memory (LSTM) building blocks to predict driver fatigue. Two parameters, Heart Rate Variability (HRV) and the Percentage of Eyelid Closure (PERCLOS) are used as an input data to the LSTM model so as to predict whether the driver will be fatigued in some future time slot. Experiments show that the system PERCLOS best using LSTM-based driver fatigue in prediction module that can achieve a true positive rate of 75% and an accuracy of 88%. Whereas fatigue prediction of the HRV best uses BPNN-based prediction module that yields a true positive rate and accuracy of 80% and 88% respectively.

Biological signals of human beings like Electroencephalograph (EEG) and Electromyography (EMG) contain plenty of motion-related information, which can reflect motion intention objectively and truly. The use of the biosignals to control mechanical equipment has been widely utilized in interactive tasks such as industrial operations, game design, hazardous operation and so forth. In this paper, we proposed an EEG-EMG based wireless collaboration technology to control the manipulator. The experiments of motion intention detection via motor imagery (MI)-EEG and surface EMG (sEMG) were conducted. Wireless manipulator control was achieved subsequently. The 2 streams of biosignals were divided into sliding windows after noise removal. Mean average value (MAV) based on Power Spectral Density (PSD) was extracted from EEG signals. Integral value of EMG (iEMG), Root Mean Square (RMS) and Wave Length (WL) were extracted from EMG signals. Support Vector Machine (SVM) and back propagation (BP) neural network were utilized to do motion classification based on features of EEG and that of EMG, respectively. The recognition accuracy of EEG signals is up to 83%, and that of EMG signals is up to 93.2%. Our EEG-EMG based wireless collaboration technology further ensures the feasibility and safety of remote operation and interaction of mechanical equipment in human-robot interaction (HRI). The actions of mechanical equipment can be fed back to the users in real time for the next adjustment, realizing friendly HRI.

Driver's drowsiness is one of the main causes of traffic accidents. It accounts for up to 20% of serious or fatal accidents on the roads. In this paper, we propose a novel sensor-based driver condition recognition to prevent drowsiness-related accidents by using Support Vector Machine (SVM). It helps to determine the driver's condition by using sensors of the wearable device. The process for the proposed warning system is as follows: First, we acquire bio-data from a PhotoPlethysmoGraphy(PPG) sensor in the device to understand the characteristics of the driver's condition. Then, the acquired data is processed through segmentation and averaging to increase classification accuracy. The processed data is used as feature vectors in the SVM for the driver's condition classification. To evaluate performance of proposed method, the following metrics were used: accuracy, error rate, precision and recall. From the calculated results, the proposed method had high accuracy of 96.3 %. Finally, we could create the driver's condition recognition model that can be applied to a system which can alert the driver.

Driver fatigue is one of the leading causes of traffic accidents. At present, fatigue driving detection has disadvant ages such as low practical application effect and high equipment requirements. This paper proposes a multi-feature point non-invasive fatigue monitoring system based on a support vector machine with a hybrid kernel function. The system detects feature points through a gradient descent tree algorithm based on a cascaded regression and calculates the eye aspect ratio (EAR) and mouth aspect ratio (MAR). The heart rate is obtained through RGB image analysis combined with Euler’s video magnification algorithm. Classify facial features to get fatigued. This paper is based on the Logistic and Radial Basis Polynomial Kernel (RBPK) function to improve the support vector machine, which has better learning and generalization. Finally, this paper uses the Driver Drowsiness Detection Dataset and the author’s dataset to test. The classification accuracy rate for a single picture is 96.92%. In summary, the system proposed in this paper has a better recognition rate for fatigue driving detection.

Early Prediction of Drug-Resistant Epilepsy using Clinical and EEG Features Based on Convolutional Neural Network

Our findings indicate the potential use of the proposed data fusion framework utilizing the graph-based features in the hybrid BCI systems by making the motor imaginary inference more accurate and more robust.

Driver fatigue is one of the main causes of traffic accidents. Many researchers believe that electroencephalography is very accurate and promising to detect driving fatigue, but so far, the research results have been variable. Through simulation experiments, this paper found when driving at night, delta band varies significantly with the degree of fatigue. The results have implications for real-time automatic detecting driver fatigue at night and are helpful for future development of driver fatigue countermeasures.

Safety driving performance estimation and alertness (SDPEA) has drawn the attention of researchers in preventing traffic accidents caused by drowsiness while driving. Psychophysiological measures, such as electroencephalogram (EEG), are accurately investigated to be robust candidates for drivers’ drowsiness evaluation. This paper presents an effective EEG-based driver drowsiness monitoring system by analyzing the changes of brain activities in a simulator driving environment. The proposed SDPEA system can translate EEG signals into drowsiness level. Firstly, Independent component analysis (ICA) is performed on EEG data to remove artifacts. Then, eight EEG-band powers-related features: beta, alpha, theta, delta, (alpha plus theta)/beta, alpha / beta, (alpha plus theta)/(alpha plus beta) and theta / beta are extracted from the preprocessed EEG signals by employing the Fast Fourier Transform (FFT). Subsequently, fisher score technique selects the most descriptive features for further classification. Finally, Support Vector Machine (SVM) is employed as a classifier to distinguish drowsiness level. Experimental results show that the quantitative driving performance can be correctly estimated through analyzing driver’s EEG signals by the SDPEA system.

Depression is a common mental disorder that seriously affects patients' social function and daily life. Its accurate diagnosis remains a big challenge in depression treatment. In this study, we used electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) and measured the whole brain EEG signals and forehead hemodynamic signals from 25 depression patients and 30 healthy subjects during the resting state. On one hand, we explored the EEG brain functional network properties, and found that the clustering coefficient and local efficiency of the delta and theta bands in patients were significantly higher than those in normal subjects. On the other hand, we extracted brain network properties, asymmetry, and brain oxygen entropy as alternative features, used a data-driven automated method to select features, and established a support vector machine model for automatic depression classification. The results showed the classification accuracy was 81.8% when using EEG features alone and increased to 92.7% when using hybrid EEG and fNIRS features. The brain network local efficiency in the delta band, hemispheric asymmetry in the theta band and brain oxygen sample entropy features differed significantly between the two groups (p < 0.05) and showed high depression distinguishing ability indicating that they may be effective biological markers for identifying depression. EEG, fNIRS and machine learning constitute an effective method for classifying depression at the individual level.

EEG (Electroencephalogram) signals are used as effective indicators for driver fatigue evaluation. Usually, EEG signals are collected by high-cost electroencephalographs which can record multi-channels of EEG data. This paper is aimed to determine interested electrodes of the relatively optimal EEG based indicator. Firstly, 16 channels of EEG data are collected and transformed into three bands (θ, α, β). Secondly, 12 types of EEG-based parameters are established and TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) is introduced to determine the optimal indicator of driver fatigue. Thirdly, the interested electrodes (Fp1, F7) are determined by MDS (multidimensional scaling). Finally, the evaluation model of driver fatigue is established with the indicator of Fp1 and F7. Experiment results verify that the efficiency has no significant difference between evaluations with two electrodes and 16 electrodes. In addition, the established evaluation model is proven to be efficient to evaluate the driver fatigue.

With the rapid growth of the number of motor vehicles, road traffic accidents are becoming more and more serious. Fatigue driving has become one of the main causes of traffic accidents. It is particularly critical to find effective solutions to fatigue driving. At present, the academic community believes that the phenomenon of fatigue driving can be controlled by means of prevention and intervention. In terms of intervention countermeasures, using sound stimulation as a wake-up method has the advantages of easy implementation and significant effect, which has become the research focus in the driving fatigue mitigation scheme. Through the establishment of a simulation driving platform, driving fatigue detection system and driving fatigue wake-up system in the laboratory, the experimental research is carried out to study the fatigue generation of drivers in different emotional types of music environment and the effectiveness of wake-up countermeasures. Three sound environments were constructed in the experiment, which were driving noise environment, positive emotional music mixed with driving noise environment and negative emotional music mixed with driving noise environment. The subjects were tested in three sound environments in turn. When entering fatigue, they were given no stimulation, positive emotional music stimulation or negative emotional music stimulation according to the experimental design. The experimental results were obtained after considering the subjective evaluation questionnaire and the experimental data analysis based on pupil diameter variation coefficient, average blink rate and other indicators. The experimental results show that when driving for a long time, the driver will enter the fatigue state regardless of whether the driving background has different emotional music. The driver's fatigue depth in the positive emotional music environment is shallower than that in the negative emotional music environment and the music-free environment; there is no significant difference in the degree of fatigue between the driver 's fatigue in the negative emotional music environment and the fatigue in the absence of music environment. In the comparison of music arousal stimuli with different emotions, the introduction of positive emotional music as arousal stimulus has the best effect.

Electroencephalogram (EEG) is commonly used for analyzing numerous psychological states of the brain. However, epileptic seizure prediction from EEG signals is quite challenging since it has more fluctuating information about the behaviour of the brain. Analyzing such long-term EEG signals to discriminate between interictal versus ictal regions is a difficult task. Also, EEG signals can be affected by noises from different sources. The proposed work presents an efficient approach based on Weighted Visibility Graph (WVG) for seizure prediction. In this work, the EEG signals are filtered to remove the artifacts due to power supply noise and then the filtered EEG time series data is segmented. The segmented time series data is converted into a complex network called WVG. This WVG inherits the dynamic characteristics of the EEG signal from which it is created. Features like mean degree, mean weighted degree and mean entropy are extracted from the WVG. These features are used to derive the essential characteristics of EEG from the WVG. Finally, classification is done using Support Vector Machine (SVM). The experiments show that the proposed system provides better performance than the existing methods in prediction of seizure in ictal as well as interictal states of EEG over the benchmark dataset.