scholarly journals Decoding of Walking Imagery and Idle State Using Sparse Representation Based on fNIRS

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Hongquan Li ◽  
Anmin Gong ◽  
Lei Zhao ◽  
Wei Zhang ◽  
Fawang Wang ◽  
...  
Keyword(s):  
Sparse Representation ◽  
Classification Accuracy ◽  
Time Windows ◽  
Support Vector ◽  
Training Method ◽  
Functional Near Infrared Spectroscopy ◽  
Idle State ◽  
Linear Discriminant ◽  
Rehabilitation Training ◽  
Active Rehabilitation

Objectives. Brain-computer interface (BCI) based on functional near-infrared spectroscopy (fNIRS) is expected to provide an optional active rehabilitation training method for patients with walking dysfunction, which will affect their quality of life seriously. Sparse representation classification (SRC) oxyhemoglobin (HbO) concentration was used to decode walking imagery and idle state to construct fNIRS-BCI based on walking imagery. Methods. 15 subjects were recruited and fNIRS signals were collected during walking imagery and idle state. Firstly, band-pass filtering and baseline drift correction for HbO signal were carried out, and then the mean value, peak value, and root mean square (RMS) of HbO and their combinations were extracted as classification features; SRC was used to identify the extracted features and the result of SRC was compared with those of support vector machine (SVM), K-Nearest Neighbor (KNN), linear discriminant analysis (LDA), and logistic regression (LR). Results. The experimental results showed that the average classification accuracy for walking imagery and idle state by SRC using three features combination was 91.55 ± 3.30%, which was significantly higher than those of SVM, KNN, LDA, and LR (86.37 ± 4.42%, 85.65 ± 5.01%, 86.43 ± 4.41%, and 76.14 ± 5.32%, respectively), and the classification accuracy of other combined features was higher than that of single feature. Conclusions. The study showed that introducing SRC into fNIRS-BCI can effectively identify walking imagery and idle state. It also showed that different time windows for feature extraction have an impact on the classification results, and the time window of 2–8 s achieved a better classification accuracy (94.33 ± 2.60%) than other time windows. Significance. The study was expected to provide a new and optional active rehabilitation training method for patients with walking dysfunction. In addition, the experiment was also a rare study based on fNIRS-BCI using SRC to decode walking imagery and idle state.

scholarly journals Fugl-Meyer hand motor imagination recognition for brain–computer interfaces using only fNIRS

2021 ◽  
Author(s):  
Chenguang Li ◽  
Hongjun Yang ◽  
Long Cheng
Keyword(s):  
Classification Accuracy ◽  
Near Infrared ◽  
Support Vector ◽  
Classification Methods ◽  
Experimental Paradigm ◽  
Functional Near Infrared Spectroscopy ◽  
Machine Method ◽  
Physiological Signal ◽  
Computer Interfaces ◽  
Mode Decomposition

AbstractAs a relatively new physiological signal of brain, functional near-infrared spectroscopy (fNIRS) is being used more and more in brain–computer interface field, especially in the task of motor imagery. However, the classification accuracy based on this signal is relatively low. To improve the accuracy of classification, this paper proposes a new experimental paradigm and only uses fNIRS signals to complete the classification task of six subjects. Notably, the experiment is carried out in a non-laboratory environment, and movements of motion imagination are properly designed. And when the subjects are imagining the motions, they are also subvocalizing the movements to prevent distraction. Therefore, according to the motor area theory of the cerebral cortex, the positions of the fNIRS probes have been slightly adjusted compared with other methods. Next, the signals are classified by nine classification methods, and the different features and classification methods are compared. The results show that under this new experimental paradigm, the classification accuracy of 89.12% and 88.47% can be achieved using the support vector machine method and the random forest method, respectively, which shows that the paradigm is effective. Finally, by selecting five channels with the largest variance after empirical mode decomposition of the original signal, similar classification results can be achieved.

scholarly journals Pilot Study on Gait Classification Using fNIRS Signals

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Hedian Jin ◽  
Chunguang Li ◽  
Jiacheng Xu
Keyword(s):  
Near Infrared ◽  
Motor Dysfunction ◽  
Assistive Device ◽  
Support Vector ◽  
Small Step ◽  
Functional Near Infrared Spectroscopy ◽  
Low Speed ◽  
Rehabilitation Training ◽  
Feature Vectors ◽  
Motion Intention

Rehabilitation training is essential for motor dysfunction patients, and the training through their subjective motion intention, comparing to passive training, is more conducive to rehabilitation. This study proposes a method to identify motion intention of different walking states under the normal environment, by using the functional near-infrared spectroscopy (fNIRS) technology. Twenty-two healthy subjects were recruited to walk with three different gaits (including small-step with low-speed, small-step with midspeed, midstep with low-speed). The wavelet packet decomposition was used to find out the main characteristic channels in different motion states, and these channels with links in frequency and space were combined to define as feature vectors. According to different permutations and combinations of all feature vectors, a library for support vector machines (libSVM) was used to achieve the best recognition model. Finally, the accuracy rate of these three walking states was 78.79%. This study implemented the classification of different states’ motion intention by using the fNIRS technology. It laid a foundation to apply the classified motion intention of different states timely, to help severe motor dysfunction patients control a walking-assistive device for rehabilitation training, so as to help them restore independent walking abilities and reduce the economic burdens on society.

scholarly journals Analysis of Different Classification Techniques for Two-Class Functional Near-Infrared Spectroscopy-Based Brain-Computer Interface

2016 ◽  
Vol 2016 ◽  
pp. 1-11 ◽  
Author(s):  
Noman Naseer ◽  
Nauman Khalid Qureshi ◽  
Farzan Majeed Noori ◽  
Keum-Shik Hong
Keyword(s):  
Infrared Spectroscopy ◽  
Discriminant Analysis ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Continuous Wave ◽  
Mental Arithmetic ◽  
Statistical Significance ◽  
Support Vector ◽  
Functional Near Infrared Spectroscopy ◽  
Linear Discriminant

We analyse and compare the classification accuracies of six different classifiers for a two-class mental task (mental arithmetic and rest) using functional near-infrared spectroscopy (fNIRS) signals. The signals of the mental arithmetic and rest tasks from the prefrontal cortex region of the brain for seven healthy subjects were acquired using a multichannel continuous-wave imaging system. After removal of the physiological noises, six features were extracted from the oxygenated hemoglobin (HbO) signals. Two- and three-dimensional combinations of those features were used for classification of mental tasks. In the classification, six different modalities, linear discriminant analysis (LDA), quadratic discriminant analysis (QDA),k-nearest neighbour (kNN), the Naïve Bayes approach, support vector machine (SVM), and artificial neural networks (ANN), were utilized. With these classifiers, the average classification accuracies among the seven subjects for the 2- and 3-dimensional combinations of features were 71.6, 90.0, 69.7, 89.8, 89.5, and 91.4% and 79.6, 95.2, 64.5, 94.8, 95.2, and 96.3%, respectively. ANN showed the maximum classification accuracies: 91.4 and 96.3%. In order to validate the results, a statistical significance test was performed, which confirmed that thepvalues were statistically significant relative to all of the other classifiers (p< 0.005) using HbO signals.

scholarly journals A Novel Fast Training Method for SVM and Its Application in Fault Diagnosis of Service Robot

2015 ◽  
Vol 11 (6) ◽  
pp. 4 ◽  
Author(s):  
Xianfeng Yuan ◽  
Mumin Song ◽  
Fengyu Zhou ◽  
Yugang Wang ◽  
Zhumin Chen
Keyword(s):  
Fault Diagnosis ◽  
Classification Accuracy ◽  
Machine Learning Algorithms ◽  
Sensor Data ◽  
Service Robot ◽  
Support Vector ◽  
Svm Classifier ◽  
Processing Unit ◽  
Training Method ◽  
Fast Training

Support Vector Machines (SVM) is a set of popular machine learning algorithms which have been successfully applied in diverse aspects, but for large training data sets the processing time and computational costs are prohibitive. This paper presents a novel fast training method for SVM, which is applied in the fault diagnosis of service robot. Firstly, sensor data are sampled under different running conditions of the robot and those samples are divided as training sets and testing sets. Secondly, the sampled data are preprocessed and the principal component analysis (PCA) model is established for fault feature extraction. Thirdly, the feature vectors are used to train the SVM classifier, which achieves the fault diagnosis of the robot. To speed up the training process of SVM, on the one hand, sample reduction is done using the proposed support vectors selection (SVS) algorithm, which can ensure good classification accuracy and generalization capability. On the other hand, we take advantage of the excellent parallel computing abilities of Graphics Processing Unit (GPU) to pre-calculate the kernel matrix, which avoids the recalculation during the cross validation process. Experimental results illustrate that the proposed method can significantly reduce the training time without decreasing the classification accuracy.

scholarly journals Use of machine learning to identify autism with natural social gaze behavior (Preprint)

2021 ◽  
Author(s):  
Zhong Zhao ◽  
Haiming Tang ◽  
Xiaobin Zhang ◽  
Xingda Qu ◽  
Jianping Lu
Keyword(s):  
Machine Learning ◽  
Eye Tracking ◽  
Classification Accuracy ◽  
Autism Spectrum ◽  
Support Vector ◽  
Gaze Behavior ◽  
Linear Discriminant ◽  
Face To Face ◽  
Area Of Interest ◽  
Social Gaze

BACKGROUND Abnormal gaze behavior is a prominent feature of the autism spectrum disorder (ASD). Previous eye tracking studies had participants watch images (i.e., picture, video and webpage), and the application of machine learning (ML) on these data showed promising results in identify ASD individuals. Given the fact that gaze behavior differs in face-to-face interaction from image viewing tasks, no study has investigated whether natural social gaze behavior could accurately identify ASD. OBJECTIVE The objective of this study was to examine whether and what area of interest (AOI)-based features extracted from the natural social gaze behavior could identify ASD. METHODS Both children with ASD and typical development (TD) were eye-tracked when they were engaged in a face-to-face conversation with an interviewer. Four ML classifiers (support vector machine, SVM; linear discriminant analysis, LDA; decision tree, DT; and random forest, RF) were used to determine the maximum classification accuracy and the corresponding features. RESULTS A maximum classification accuracy of 84.62% were achieved with three classifiers (LDA, DT and RF). Results showed that the mouth, but not the eyes AOI, was a powerful feature in detecting ASD. CONCLUSIONS Natural gaze behavior could be leveraged to identify ASD, suggesting that ASD might be objectively screened with eye tracking technology in everyday social interaction. In addition, the comparison between our and previous findings suggests that eye tracking features that could identify ASD might be culture dependent and context sensitive.

scholarly journals Open-Access fNIRS Dataset for Classification of Unilateral Finger- and Foot-Tapping

Electronics ◽  
2019 ◽  
Vol 8 (12) ◽  
pp. 1486
Author(s):  
SuJin Bak ◽  
Jinwoo Park ◽  
Jaeyoung Shin ◽  
Jichai Jeong
Keyword(s):  
Open Access ◽  
Classification Accuracy ◽  
Near Infrared ◽  
Finger Tapping ◽  
Support Vector ◽  
Functional Near Infrared Spectroscopy ◽  
Svm Classification ◽  
Right Hand ◽  
Foot Tapping ◽  
Concentration Changes

Numerous open-access electroencephalography (EEG) datasets have been released and widely employed by EEG researchers. However, not many functional near-infrared spectroscopy (fNIRS) datasets are publicly available. More fNIRS datasets need to be freely accessible in order to facilitate fNIRS studies. Toward this end, we introduce an open-access fNIRS dataset for three-class classification. The concentration changes of oxygenated and reduced hemoglobin were measured, while 30 volunteers repeated each of the three types of overt movements (i.e., left- and right-hand unilateral complex finger-tapping, foot-tapping) for 25 times. The ternary support vector machine (SVM) classification accuracy obtained using leave-one-out cross-validation was estimated at 70.4% ± 18.4% on average. A total of 21 out of 30 volunteers scored a superior binary SVM classification accuracy (left-hand vs. right-hand finger-tapping) of over 80.0%. We believe that the introduced fNIRS dataset can facilitate future fNIRS studies.

scholarly journals Detection of primary RGB colors projected on a screen using fNIRS

2017 ◽  
Vol 10 (03) ◽  
pp. 1750006 ◽  
Author(s):  
Xiaolong Liu ◽  
Keum-Shik Hong
Keyword(s):  
Infrared Spectroscopy ◽  
Visual Cortex ◽  
Discriminant Analysis ◽  
Near Infrared Spectroscopy ◽  
Classification Accuracy ◽  
Near Infrared ◽  
Functional Near Infrared Spectroscopy ◽  
Linear Discriminant ◽  
Stimulation Period ◽  
Skewness And Kurtosis

In this study, functional near-infrared spectroscopy (fNIRS) is utilized to measure the hemodynamic responses (HRs) in the visual cortex of 14 subjects (aged 22–34 years) viewing the primary red, green, and blue (RGB) colors displayed on a white screen by a beam projector. The spatiotemporal characteristics of their oxygenated and deoxygenated hemoglobins (HbO and HbR) in the visual cortex are measured using a 15-source and 15-detector optode configuration. To see whether the activation maps upon RGB-color stimuli can be distinguished or not, the [Formula: see text]-values of individual channels are averaged over 14 subjects. To find the best combination of two features for classification, the HRs of activated channels are averaged over nine trials. The HbO mean, peak, slope, skewness and kurtosis values during 2–7[Formula: see text]s window for a given 10[Formula: see text]s stimulation period are analyzed. Finally, the linear discriminant analysis (LDA) for classifying three classes is applied. Individually, the best classification accuracy obtained with slope-skewness features was 74.07% (Subject 1), whereas the best overall over 14 subjects was 55.29% with peak-skewness combination. Noting that the chance level of 3-class classification is 33.33%, it can be said that RGB colors can be distinguished. The overall results reveal that fNIRS can be used for monitoring purposes of the HR patterns in the human visual cortex.

scholarly journals Evaluation of Methods for Estimating Fractal Dimension in Motor Imagery-Based Brain Computer Interface

2011 ◽  
Vol 2011 ◽  
pp. 1-8 ◽  
Author(s):  
Chu Kiong Loo ◽  
Andrews Samraj ◽  
Gin Chong Lee
Keyword(s):  
Feature Extraction ◽  
Fractal Dimension ◽  
Motor Imagery ◽  
Classification Accuracy ◽  
Brain Activity ◽  
Brain Computer Interface ◽  
Computer Interface ◽  
Estimation Methods ◽  
Support Vector ◽  
Linear Discriminant

A brain computer interface BCI enables direct communication between a brain and a computer translating brain activity into computer commands using preprocessing, feature extraction, and classification operations. Feature extraction is crucial, as it has a substantial effect on the classification accuracy and speed. While fractal dimension has been successfully used in various domains to characterize data exhibiting fractal properties, its usage in motor imagery-based BCI has been more recent. In this study, commonly used fractal dimension estimation methods to characterize time series Katz's method, Higuchi's method, rescaled range method, and Renyi's entropy were evaluated for feature extraction in motor imagery-based BCI by conducting offline analyses of a two class motor imagery dataset. Different classifiers fuzzy k-nearest neighbours FKNN, support vector machine, and linear discriminant analysis were tested in combination with these methods to determine the methodology with the best performance. This methodology was then modified by implementing the time-dependent fractal dimension TDFD, differential fractal dimension, and differential signals methods to determine if the results could be further improved. Katz's method with FKNN resulted in the highest classification accuracy of 85%, and further improvements by 3% were achieved by implementing the TDFD method.

Evidence-Based Combination of Weighted Classifiers Approach for Epileptic Seizure Detection using EEG Signals

2012 ◽  
Vol 3 (2) ◽  
pp. 27-44
Author(s):  
Abduljalil Mohamed ◽  
Khaled Bashir Shaban ◽  
Amr Mohamed
Keyword(s):  
Epileptic Seizure ◽  
Classification Accuracy ◽  
Evidence Theory ◽  
Seizure Detection ◽  
Support Vector ◽  
Brain State ◽  
Eeg Signals ◽  
Epileptic Seizure Detection ◽  
Linear Discriminant ◽  
Detection Techniques

Different brain states and conditions can be captured by electroencephalogram (EEG) signals. EEG-based epileptic seizure detection techniques often reduce these signals into sets of discriminant features. In this work, an evidence theory-based approach for epileptic detection, using several classifiers, is proposed. Within the framework of the evidence theory, each of these classifiers is considered a source of information and given a certain weight based on both its overall classification accuracy as well as its precision rate for the respective brain state. These sources are fused using the Dempster’s rule of combination. Experimental work is done where five time domain features are obtained from EEG signals and used by a set classifiers, namely, Bayesian, K-nearest neighbor, neural network, linear discriminant analysis, and support vector machine classifiers. Higher classification accuracy of 89.5% is achieved, compared to 75.07% and 87.71% accuracy obtained from the worst and best used classifiers.

scholarly journals Classifier Performance Evaluation in Wrist and Finger Movement Fitting Task Based on Forearm HD-sEMG

2021 ◽  
Author(s):  
Haiqiang Duan ◽  
Chenyun Dai ◽  
Wei Chen
Keyword(s):  
Classification Accuracy ◽  
Nearest Neighbor ◽  
Hand Movement ◽  
The Other ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Linear Superposition ◽  
Semg Signal ◽  
Linear Discriminant ◽  
Fitting In

Abstract Background: The transmission of human body movements to other devices through wearable smart bracelets have attracted more and more attentions in the field of human-machine interface (HMI) applications. However, due to the limitation of the collection range of wearable bracelets, it is necessary to study the relationship between the superposition of wrist and finger motion and their cooperative motion to simplify the collection system of the device.Methods: The multi-channel high-density surface electromyogram (HD-sEMG) signal has high spatial resolution and can improve the accuracy of multi-channel fitting. In this study, we quantified the HD-sEMG forearm spatial activation features of 256 channels of hand movement, and performed a linear fitting of the quantified features of fingers and wrist movements to verify the linear superposition relationship between fingers and wrist cooperative movements and their independent movements. The most important thing is to classify and predict the results of the fitting and the actual measured fingers and wrist cooperative actions by four commonly used classifiers: Linear Discriminant Analysis (LDA) ,K-Nearest Neighbor (KNN) ,Support Vector Machine (SVM) and Random Forest (RF), and evaluate the performance of the four classifiers in gesture fitting in detail according to the classification results.Results: In a total of 12 kinds of synthetic gesture actions, in the three cases where the number of fitting channels was selected as 8, 32 and 64, four classifiers of LDA, SVM, RF and KNN are used for classification prediction. When the number of fitting channels was 8, the prediction accuracy of LDA classifier was 99.70%, the classification accuracy of KNN was 99.40%, the classification accuracy of SVM was 99.20%, and the classification accuracy of RF was 93.75%. When the number of fitting channels was 32, the accuracy of LDA was 98.51%, the classification accuracy of KNN was 97.92%, the accuracy of SVM is 96.73%, and the accuracy of RF was 86.61%. When the number of fitting channels is 64, the accuracy of LDA is 95.83%, the classification accuracy of KNN is 91.67%, the accuracy of SVM is 86.90%, and the accuracy of RF is 83.30%.Conclusion: It can be seen from the results that when the number of fitting channels is 8, the classification accuracy of the three classifiers of LDA, KNN and SVM is basically the same, but the time-consuming of SVM is very small. When the amount of data is large, the priority should be selected SVM as the classifier. When the number of fitting channels increases, the classification accuracy of the LDA classifier will be higher than the other three classifiers, so the LDA classifier should be more appropriate. The classification accuracy of the RF classifier in this type of problem has always been far lower than the other three classifiers, so it is not recommended to use the RF classifier as a classifier for gesture stacking related work.

Sign in / Sign up

Export Citation Format

Share Document