scholarly journals Drowsiness discrimination in an overnight driving simulation on the basis of RR and QT intervals

2017 ◽  
Vol 3 (2) ◽  
pp. 563-567
Author(s):  
Christian Heinze ◽  
Constantin Hütterer ◽  
Thomas Schnupp ◽  
Gustavo Lenis ◽  
Martin Golz
Keyword(s):  
Empirical Distribution ◽  
Driving Simulation ◽  
Classification Error ◽  
Support Vector ◽  
Qt Intervals ◽  
Power Spectral ◽  
Vector Machines ◽  
Power Spectral Densities ◽  
The Mean ◽  
Driver Sleepiness

AbstractWe examined if ECG-based features are discrimi-native towards drowsiness. Twenty-five volunteers (19–32 years) completed 7×40 minutes of monotonous overnight driving simulation, designed to induce drowsiness. ECG (512 s-1) was recorded continuously; subjective ratings of drowsiness on the Karolinska sleepiness scale (KSS) were polled every five minutes. ECG recordings were divided into 5-min segments, each associated with the mean of one self- and two observer-KSS ratings. Those mean KSS values were binarized to obtain two classes not drowsy and drowsy. The Q-, R- and T-waves in the recordings were detected; R-peak positions were manually reviewed; the Q- and T-detection method was tested against the manual annotations of Physio-net’s QT database. Power spectral densities of RR intervals (RR-PSD) and quantiles of the empirical distribution of heart-rate corrected QTc intervals were estimated. Support-vector machines and random-holdout cross-validation were used for the estimation of the classification error. Using either RR-PSD or QTc features yielded mean test errors of 79.3 ± 0.3 % and 82.7 ± 0.5 %, respectively. Merging RR and QTc features improved the accuracy to 88.3 ± 0.2 %. QTc intervals of the class drowsy were generally prolonged com-pared to not drowsy. Our findings indicate that the inclusion of QT intervals contribute to the discrimination of driver sleepiness.

scholarly journals EOG feature relevance determination for microsleep detection

2017 ◽  
Vol 3 (2) ◽  
pp. 815-818
Author(s):  
Martin Golz ◽  
Sebastian Wollner ◽  
David Sommer ◽  
Sebastian Schnieder
Keyword(s):  
Cross Correlation ◽  
Feature Reduction ◽  
Support Vector ◽  
Spectral Bands ◽  
Automatic Relevance Determination ◽  
Signal Features ◽  
Power Spectral ◽  
Vector Machines ◽  
Power Spectral Densities ◽  
Feature Relevance

AbstractAutomatic relevance determination (ARD) was applied to two-channel EOG recordings for microsleep event (MSE) recognition. 10 s immediately before MSE and also before counterexamples of fatigued, but attentive driving were analysed. Two type of signal features were extracted: the maximum cross correlation (MaxCC) and logarithmic power spectral densities (PSD) averaged in spectral bands of 0.5 Hz width ranging between 0 and 8 Hz. Generalised learn-ing vector quantisation (GRLVQ) was used as ARD method to show the potential of feature reduction. This is compared to support-vector machines (SVM), in which the feature reduction plays a much smaller role. Cross validation yielded mean normalised relevancies of PSD features in the range of 1.6 - 4.9 % and 1.9 - 10.4 % for horizontal and vertical EOG, respectively. MaxCC relevancies were 0.002 - 0.006 % and 0.002 - 0.06 %, respectively. This shows that PSD features of vertical EOG are indispensable, whereas MaxCC can be neglected. Mean classification accuracies were estimated at 86.6±b 1.3 % and 92.3±b 0.2 % for GRLVQ and SVM, respec-tively. GRLVQ permits objective feature reduction by inclu-sion of all processing stages, but is not as accurate as SVM.

scholarly journals EOG feature relevance determination for microsleep detection

2017 ◽  
Vol 3 (2) ◽  
pp. 261-264
Author(s):  
Martin Golz ◽  
Sebastian Wollner ◽  
David Sommer ◽  
Sebastian Schnieder
Keyword(s):  
Cross Correlation ◽  
Feature Reduction ◽  
Support Vector ◽  
Spectral Bands ◽  
Automatic Relevance Determination ◽  
Signal Features ◽  
Power Spectral ◽  
Vector Machines ◽  
Power Spectral Densities ◽  
Feature Relevance

AbstractAutomatic relevance determination (ARD) was applied to two-channel EOG recordings for microsleep event (MSE) recognition. 10 s immediately before MSE and also before counterexamples of fatigued, but attentive driving were analysed. Two type of signal features were extracted: the maximum cross correlation (MaxCC) and logarithmic power spectral densities (PSD) averaged in spectral bands of 0.5 Hz width ranging between 0 and 8 Hz. Generalised learn-ing vector quantisation (GRLVQ) was used as ARD method to show the potential of feature reduction. This is compared to support-vector machines (SVM), in which the feature reduction plays a much smaller role. Cross validation yielded mean normalised relevancies of PSD features in the range of 1.6 – 4.9 % and 1.9 – 10.4 % for horizontal and vertical EOG, respectively. MaxCC relevancies were 0.002 – 0.006 % and 0.002 – 0.06 %, respectively. This shows that PSD features of vertical EOG are indispensable, whereas MaxCC can be neglected. Mean classification accuracies were estimated at 86.6±b 1.3 % and 92.3±b 0.2 % for GRLVQ and SVM, respectively. GRLVQ permits objective feature reduction by inclusion of all processing stages, but is not as accurate as SVM.

Generalized Models for the Classification of Abnormal Movements in Daily Life and its Applicability to Epilepsy Convulsion Recognition

2016 ◽  
Vol 26 (06) ◽  
pp. 1650037 ◽  
Author(s):  
José R. Villar ◽  
Paula Vergara ◽  
Manuel Menéndez ◽  
Enrique de la Cal ◽  
Víctor M. González ◽  
...  
Keyword(s):  
Fuzzy Systems ◽  
Classification Error ◽  
Support Vector ◽  
False Alarms ◽  
Abnormal Movements ◽  
Vector Machines ◽  
Clonic Convulsions ◽  
Event Based ◽  
Validation Scheme

The identification and the modeling of epilepsy convulsions during everyday life using wearable devices would enhance patient anamnesis and monitoring. The psychology of the epilepsy patient penalizes the use of user-driven modeling, which means that the probability of identifying convulsions is driven through generalized models. Focusing on clonic convulsions, this pre-clinical study proposes a method for generating a type of model that can evaluate the generalization capabilities. A realistic experimentation with healthy participants is performed, each with a single 3D accelerometer placed on the most affected wrist. Unlike similar studies reported in the literature, this proposal makes use of [Formula: see text] cross-validation scheme, in order to evaluate the generalization capabilities of the models. Event-based error measurements are proposed instead of classification-error measurements, to evaluate the generalization capabilities of the model, and Fuzzy Systems are proposed as the generalization modeling technique. Using this method, the experimentation compares the most common solutions in the literature, such as Support Vector Machines, [Formula: see text]-Nearest Neighbors, Decision Trees and Fuzzy Systems. The event-based error measurement system records the results, penalizing those models that raise false alarms. The results showed the good generalization capabilities of Fuzzy Systems.

Parameters Selection and Application of LS-SVM Based on Chaotic Ant Swarm Algorithm

2011 ◽  
Vol 204-210 ◽  
pp. 423-426
Author(s):  
Chun Li Xie ◽  
Dan Dan Zhao ◽  
Juan Wang ◽  
Cheng Shao
Keyword(s):  
Support Vector Machines ◽  
Nonlinear Systems ◽  
Fitness Function ◽  
Selection Method ◽  
Support Vector ◽  
Mean Square ◽  
Parameters Selection ◽  
Vector Machines ◽  
The Mean ◽  
Simulation Results

Parameters selection plays an important role for the performance of least squares support vector machines (LS-SVM). In this paper, a novel parameters selection method for LS-SVM is presented based on chaotic ant swarm (CAS) algorithm. Using this method, the optimization model is established, within which the fitness function is the mean square error (MSE) index, and the constraints are the ranges of the designing parameters. The proposed method is used in the identification for inverse model of the nonlinear systems, and simulation results are given to show the efficiency.

scholarly journals Image analysis for taxonomic identification of Javanese butterflies

2018 ◽  
Author(s):  
Saskia de Vetter ◽  
Rutger Vos
Keyword(s):  
Image Analysis ◽  
Image Features ◽  
Support Vector ◽  
Good Prediction ◽  
Vector Machines ◽  
Characteristic Image ◽  
Out Of Sample ◽  
Single Dataset ◽  
The Mean ◽  
Selection Of

SummaryTaxonomic experts classify millions of specimens, but this is very time-consuming and therefore expensive. Image analysis is a way to automate identification and was previously done at Naturalis Biodiversity Center for slipper orchids (Cypripedioideae) by the program ‘OrchID’. This program operated by extracting a pre-defined number of features from images, and these features were used to train artificial neural networks (ANN) to classify out-of-sample images. This program was extended to work for a collection of Javanese butterflies, donated to Naturalis by the Van Groenendael-Krijger Foundation. Originally, for the orchids, an image was divided into a pre-defined number of horizontal and vertical bins and the mean blue-green-red values of each bin were calculated (BGR method) to obtain image features. In the extended implementation, characteristic image features were extracted using the SURF algorithm implemented in OpenCV and clustered with the BagOfWords method (SURF-BOW method). In addition, a combination of BGR- and SURF-BOW was implemented to extract both types of features in a single dataset (BGR-SURF method). A selection of the butterfly and orchid images was made to create datasets with at least 5 and at most 50 specimens per species. The SURF-BOW and BGR-SURF methods were applied to both selected datasets, and the original BGR method was applied to the selected butterfly dataset. PCA plots were made to inspect visually how well the applied methods discriminated among the species. For the butterflies, both genus and species appeared to cluster together in the plots of the SURF-BOW method. However, no obvious clustering was noticeable for the orchid plots. The performance of the ANNs was validated by a stratified k-fold cross validation. For the butterflies, the BGR-SURF method scored best with an accuracy of 77%, against 71% for the SURF-BOW method and 66% for the BGR method, all for chained genus and species prediction with k = 10. The new methods could not improve the accuracy of the orchid classification with k = 10, which was 75% on genus, 52% on genus and section and 48% on genus, section and species in the original framework and now less than 25% for all. The validation results also showed that at least about 15 specimens per species were necessary for a good prediction with the SURF-BOW method. The BGR-SURF method was found to be the best of these methods for butterflies, but the original BGR method was best for the slipper orchids. In the future these methods may be tested with other datasets, for example with mosquitoes. In addition, other classifiers may be tested for better performance, like support vector machines.

Flaw Classification of Flexible Copper Clad-Laminated Sheet

2012 ◽  
Vol 579 ◽  
pp. 52-59
Author(s):  
Yih Chih Chiou ◽  
Yu Teng Liang
Keyword(s):  
Error Rate ◽  
Copper Foil ◽  
Classification Error ◽  
Support Vector ◽  
Classification Rate ◽  
Classification Error Rate ◽  
Vector Machines ◽  
Flaw Classification ◽  
Misclassification Rates

This study investigated the classification error rate of eleven flaws commonly occurred in copper foil. The goal was to online identify the type of the flaw being discovered in order to trace the source of the flaw and act correspondingly. The misclassification rates of four popular classifiers were investigated and compared. The results indicated that the best classification rate can be obtained by choosing Support Vector Machines as the classifier and employing all the ten features. The resulting low classification error rate of 4.41% proved the effectiveness of the derived classifier as well as the suitability of the chosen features.

scholarly journals Beyond the Hype: Deep Neural Networks Outperform Established Methods Using A ChEMBL Bioactivity Benchmark Set

2017 ◽  
Author(s):  
Eelke B. Lenselink ◽  
Niels ten Dijke ◽  
Brandon Bongers ◽  
George Papadatos ◽  
Herman W.T. van Vlijmen ◽  
...  
Keyword(s):  
Neural Networks ◽  
Logistic Regression ◽  
Support Vector Machines ◽  
Random Forests ◽  
Deep Neural Networks ◽  
Direct Result ◽  
Support Vector ◽  
Temporal Validation ◽  
Vector Machines ◽  
The Mean

AbstractThe increase of publicly available bioactivity data in recent years has fueled and catalyzed research in chemogenomics, data mining, and modeling approaches. As a direct result, over the past few years a multitude of different methods have been reported and evaluated, such as target fishing, nearest neighbor similarity-based methods, and Quantitative Structure Activity Relationship (QSAR)-based protocols. However, such studies are typically conducted on different datasets, using different validation strategies, and different metrics.In this study, different methods were compared using one single standardized dataset obtained from ChEMBL, which is made available to the public, using standardized metrics (BEDROC and Matthews Correlation Coefficient). Specifically, the performance of Naive Bayes, Random Forests, Support Vector Machines, Logistic Regression, and Deep Neural Networks was assessed using QSAR and proteochemometric (PCM) methods. All methods were validated using both a random split validation and a temporal validation, with the latter being a more realistic benchmark of expected prospective execution.Deep Neural Networks are the top performing classifiers, highlighting the added value of Deep Neural Networks over other more conventional methods. Moreover, the best method (‘DNN_PCM’) performed significantly better at almost one standard deviation higher than the mean performance. Furthermore, Multi task and PCM implementations were shown to improve performance over single task Deep Neural Networks. Conversely, target prediction performed almost two standard deviations under the mean performance. Random Forests, Support Vector Machines, and Logistic Regression performed around mean performance. Finally, using an ensemble of DNNs, alongside additional tuning, enhanced the relative performance by another 27% (compared with unoptimized DNN_PCM).Here, a standardized set to test and evaluate different machine learning algorithms in the context of multitask learning is offered by providing the data and the protocols.

scholarly journals A Study on Sensitive Bands of EEG Data under Different Mental Workloads

Algorithms ◽  
2019 ◽  
Vol 12 (7) ◽  
pp. 145
Author(s):  
Hongquan Qu ◽  
Zhanli Fan ◽  
Shuqin Cao ◽  
Liping Pang ◽  
Hao Wang ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Support Vector ◽  
Svm Classifier ◽  
Eeg Signals ◽  
Power Spectral ◽  
Eeg Data ◽  
Vector Machines ◽  
Full Band ◽  
Electroencephalogram Eeg

Electroencephalogram (EEG) signals contain a lot of human body performance information. With the development of the brain–computer interface (BCI) technology, many researchers have used the feature extraction and classification algorithms in various fields to study the feature extraction and classification of EEG signals. In this paper, the sensitive bands of EEG data under different mental workloads are studied. By selecting the characteristics of EEG signals, the bands with the highest sensitivity to mental loads are selected. In this paper, EEG signals are measured in different load flight experiments. First, the EEG signals are preprocessed by independent component analysis (ICA) to remove the interference of electrooculogram (EOG) signals, and then the power spectral density and energy are calculated for feature extraction. Finally, the feature importance is selected based on Gini impurity. The classification accuracy of the support vector machines (SVM) classifier is verified by comparing the characteristics of the full band with the characteristics of the β band. The results show that the characteristics of the β band are the most sensitive in EEG data under different mental workloads.

scholarly journals The role of expert evaluation for microsleep detection

2015 ◽  
Vol 1 (1) ◽  
pp. 92-95
Author(s):  
M. Golz ◽  
A. Schenka ◽  
D. Sommer ◽  
B. Geißler ◽  
A. Muttray
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Driving Simulation ◽  
Support Vector ◽  
Expert Evaluation ◽  
Simulation Studies ◽  
Independent Information ◽  
Vector Machines ◽  
Adaptive Pattern

AbstractRecently, it has been shown by overnight driving simulation studies that microsleep density is the only known sleepiness indicator which rapidly increases within a few seconds immediately before sleepiness related crashes. This indicator is based solely on EEG and EOG and subsequent adaptive pattern recognition. Accurate microsleep recognition is very important for the performance of this sleepiness indicator. The question is whether expensive evaluations of microsleep events by a) experts are necessary or b) non-experts provide sufficient evaluations. Based on 11,114 microsleep events in case a) and 12,787 in case b) recognition accuracies were investigated utilizing (i) artificial neural networks and (ii) support-vector machines. Cross validated classification accuracies ranged between 92.2 % for (i,b) and 99.3 % for (ii, a). It is concluded that expert evaluations are very important to provide independent information for detecting microsleep.

Sign in / Sign up

Export Citation Format

Share Document