Subject-specific Adaptation Research Articles

Deep learning classification of EEG-based BCI monitoring of the attempted arm and hand movements.

The primary objective of this research is to improve the average classification performance for specific movements in patients with cervical spinal cord injury (SCI). The study utilizes a low-frequency multi-class electroencephalography (EEG) dataset from Graz University of Technology. The research combines convolutional neural network (CNN) and long-short-term memory (LSTM) architectures to uncover neural correlations between temporal and spatial aspects of the EEG signals associated with attempted arm and hand movements. To achieve this, three different methods are used to select relevant features, and the proposed model's robustness against variations in the data is validated using 10-fold cross-validation (CV). The research also investigates subject-specific adaptation in an online paradigm, extending movement classification proof-of-concept. The combined CNN-LSTM model, enhanced by three feature selection methods, demonstrates robustness with a mean accuracy of 75.75 % and low standard deviation (+/- 0.74 %) in 10-fold cross-validation, confirming its reliability. In summary, this research aims to make valuable contributions to the field of neuro-technology by developing EEG-controlled assistive devices using a generalized brain-computer interface (BCI) and deep learning (DL) framework. The focus is on capturing high-level spatiotemporal features and latent dependencies to enhance the performance and usability of EEG-based assistive technologies.

Subject-independent EEG classification based on a hybrid neural network.

A brain-computer interface (BCI) based on the electroencephalograph (EEG) signal is a novel technology that provides a direct pathway between human brain and outside world. For a traditional subject-dependent BCI system, a calibration procedure is required to collect sufficient data to build a subject-specific adaptation model, which can be a huge challenge for stroke patients. In contrast, subject-independent BCI which can shorten or even eliminate the pre-calibration is more time-saving and meets the requirements of new users for quick access to the BCI. In this paper, we design a novel fusion neural network EEG classification framework that uses a specially designed generative adversarial network (GAN), called a filter bank GAN (FBGAN), to acquire high-quality EEG data for augmentation and a proposed discriminative feature network for motor imagery (MI) task recognition. Specifically, multiple sub-bands of MI EEG are first filtered using a filter bank approach, then sparse common spatial pattern (CSP) features are extracted from multiple bands of filtered EEG data, which constrains the GAN to maintain more spatial features of the EEG signal, and finally we design a convolutional recurrent network classification method with discriminative features (CRNN-DF) to recognize MI tasks based on the idea of feature enhancement. The hybrid neural network proposed in this study achieves an average classification accuracy of 72.74 ± 10.44% (mean ± std) in four-class tasks of BCI IV-2a, which is 4.77% higher than the state-of-the-art subject-independent classification method. A promising approach is provided to facilitate the practical application of BCI.

Subject-specific compliance control of an upper-limb bilateral robotic system

This paper proposes a new compliance control strategy on a robot-assisted bilateral upper limb rehabilitation system. The robotic compliance regulation was achieved by modifying predefined training trajectories in real time, based on measured human–robot interaction force and human users’ position within subject-specific workspace. Experimental data were collected from healthy subjects, and indicate that human users can follow predefined training trajectories well under real-time compliance variation, with the maximum normalized root mean square error (NRMSE) value no greater than 1.44%. Preliminary findings are encouraging, which demonstrates the availability of the proposed subject-specific compliance adaptation strategy, and the potential with enhanced training safety and efficacy. Future work will consider conducting a direct comparison between a bilateral upper limb rehabilitation device (BULReD)-assisted compliance training with or without subject-specific adaptation on a large sample of participants with upper limb disabilities.

Motor Imagery Classification via Temporal Attention Cues of Graph Embedded EEG Signals.

Motor imagery classification from EEG signals is essential for motor rehabilitation with a Brain-Computer Interface (BCI). Most current works on this issue require a subject-specific adaptation step before applied to a new user. Thus the research of directly extending a pre-trained model to new users is particularly desired and indispensable. As brain dynamics fluctuate considerably across different subjects, it is challenging to design practical hand-crafted features based on prior knowledge. Regarding this gap, this paper proposes a Graph-based Convolutional Recurrent Attention Model (G-CRAM) to explore EEG features across different subjects for motor imagery classification. A graph structure is first developed to represent the positioning information of EEG nodes. Then a convolutional recurrent attention model learns EEG features from both spatial and temporal dimensions and emphasizes on the most distinguishable temporal periods. We evaluate the proposed approach on two benchmark EEG datasets of motor imagery classification on the subject-independent testing. The results show that the G-CRAM achieves superior performance to state-of-the-art methods regarding recognition accuracy and ROC-AUC. Furthermore, model interpretation studies reveal the learning process of different neural network components and demonstrate that the proposed model can extract detailed features efficiently.

A Graph-Based Hierarchical Attention Model for Movement Intention Detection from EEG Signals.

An EEG-based Brain-Computer Interface (BCI) is a system that enables a user to communicate with and intuitively control external devices solely using the user's intentions. Current EEG-based BCI research usually involves a subject-specific adaptation step before a BCI system is ready to be employed by a new user. However, the subject-independent scenario, in which a well-trained model can be directly applied to new users without pre-calibration, is particularly desirable yet rarely explored. Considering this critical gap, our focus in this paper is the subject-independent scenario of EEG-based human intention recognition. We present a G raph-based H ierarchical A ttention M odel (G-HAM) that utilizes the graph structure to represent the spatial information of EEG sensors and the hierarchical attention mechanism to focus on both the most discriminative temporal periods and EEG nodes. Extensive experiments on a large EEG dataset containing 105 subjects indicate that our model is capable of exploiting the underlying invariant EEG patterns across different subjects and generalizing the patterns to new subjects with better performance than a series of state-of-the-art and baseline approaches.

A Convolutional Recurrent Attention Model for Subject-Independent EEG Signal Analysis

The electroencephalogram (EEG) signal is a medium to realize a brain–computer interface (BCI) system due to its zero clinical risk and portable acquisition devices. Current EEG-based BCI research usually requires a subject-specific adaptation step before a BCI can be employed by a new user. In contrast, the subject-independent scenario, where a well trained model can be directly applied to new users without precalibration, is particularly desired. Considering this critical gap, the focus in this letter is developing an effective EEG signal analysis adaptively applied to subject-independent settings. We present a convolutional recurrent attention model ( CRAM ) that utilizes a convolutional neural network to encode the high-level representation of EEG signals and a recurrent attention mechanism to explore the temporal dynamics of the EEG signals as well as to focus on the most discriminative temporal periods. Extensive experiments on a benchmark multiclass EEG dataset containing four movement intentions indicate that the proposed model is capable of exploiting the underlying invariant EEG patterns across different subjects and generalizing the patterns to new subjects with better performance than a series of state-of-the-art and baseline approaches by at least eight percentage points. The implementation code is made publicly available. 1 1 https://github.com/dalinzhang/CRAM .

Context-Dependent Risk Aversion: A Model-Based Approach.

Most research on risk aversion in behavioral science with human subjects has focused on a component of risk aversion that does not adapt itself to context. More recently, studies have explored risk aversion adaptation to changing circumstances in sequential decision-making tasks. It is an open question whether one can identify evidence, at the single subject level, for such risk aversion adaptation. We conducted a behavioral experiment on human subjects, using a sequential decision making task. We developed a model-based approach for estimating the adaptation of risk-taking behavior with single-trial resolution by modeling a subject's goals and internal representation of task contingencies. Using this model-based approach, we estimated the subject-specific adaptation of risk aversion depending on the current task context. We found striking inter-subject variations in the adaptation of risk-taking behavior. We show that these differences can be explained by differences in subjects' internal representations of task contingencies and goals. We discuss that the proposed approach can be adapted to a wide range of experimental paradigms and be used to analyze behavioral measures other than risk aversion.

An end-to-end deep learning approach to MI-EEG signal classification for BCIs

Goal: To develop and implement a Deep Learning (DL) approach for an electroencephalogram (EEG) based Motor Imagery (MI) Brain-Computer Interface (BCI) system that could potentially be used to improve the current stroke rehabilitation strategies.Method: The DL model is using Convolutional Neural Network (CNN) layers for learning generalized features and dimension reduction, while a conventional Fully Connected (FC) layer is used for classification. Together they build a unified end-to-end model that can be applied to raw EEG signals. This previously proposed model was applied to a new set of data to validate its robustness against data variations. Furthermore, it was extended by subject-specific adaptation. Lastly, an analysis of the learned filters provides insights into how such a model derives a classification decision.Results: The selected global classifier reached 80.38%, 69.82%, and 58.58% mean accuracies for datasets with two, three, and four classes, respectively, validated using 5-fold crossvalidation. As a novel approach in this context, transfer learning was used to adapt the global classifier to single individuals improving the overall mean accuracy to 86.49%, 79.25%, and 68.51%, respectively. The global models were trained on 3s segments of EEG data from different subjects than they were tested on, which proved the generalization performance of the model.Conclusion: The results are comparable with the reported accuracy values in related studies and the presented model outperforms the results in the literature on the same underlying data. Given that the model can learn features from data without having to use specialized feature extraction methods, DL should be considered as an alternative to established EEG classification methods, if enough data is available.

A topological approach to delineation and arrhythmic beats detection in unprocessed long-term ECG signals

Background and objectiveArrhythmias are one of the most common symptoms of cardiac failure. They are usually diagnosed using ECG recordings, particularly long ambulatory recordings (AECG). These recordings are tedious to interpret by humans due to their extent (up to 48 h) and the relative scarcity of arrhythmia events. This makes automated systems for detecting various AECG anomalies indispensable. In this work we present a novel procedure based on topological principles (Morse theory) for detecting arrhythmic beats in AECG. It works in nearly real-time (delayed by a 14 s window), and can be applied to raw (unprocessed) ECG signals. MethodsThe procedure is based on a subject-specific adaptation of the one-dimensional discrete Morse theory (ADMT), which represents the signal as a sequence of its most important extrema. The ADMT algorithm is applied twice; for low-amplitude, high-frequency noise removal, and for detection of the characteristic waves of individual ECG beats. The waves are annotated using the ADMT algorithm and template matching. The annotated beats are then compared to the adjacent beats with two measures of similarity: the distance between two beats, and the difference in shape between them. The two measures of similarity are used as inputs to a decision tree algorithm that classifies the beats as normal or abnormal. The classification performance is evaluated with the leave-one-record-out cross-validation method. ResultsOur approach was tested on the MIT-BIH database, where it exhibited a classification accuracy of 92.73%, a sensitivity of 73.35%, a specificity of 96.70%, a positive predictive value of 88.01%, and a negative predictive value of 95.73%. ConclusionsCompared to related studies, our algorithm requires less preprocessing while retaining the capability to detect and classify beats in almost real-time. The algorithm exhibits a high degree of accuracy in beats detection and classification that are at least comparable to state-of-the-art methods.

A three-stage trajectory generation method for robot-assisted bilateral upper limb training with subject-specific adaptation

Robot-assisted bilateral upper limb training typically requires two robotic devices that work cooperatively with a human user. The determination of appropriate movement trajectories is essential to avoid interference of the robotic systems and ensure an appropriate intersecting workspace with user limbs. This paper proposes a new three-stage trajectory generation method for bilateral upper limb training using interference analysis. These three stages include workspace analysis of robots and human hands, trajectory generation within the intersecting workspace, and interference analysis for training safety verification. This trajectory generation method is also implemented with subject-specific adaptation based on anthropometry of an individual. Experiments were conducted on seven healthy subjects with a variety of body sizes. All participants gave positive feedback on the suitability of the predefined training trajectories, and no robot interference was detected. This three-stage method provides guidelines for the standardization of robot-assisted bilateral upper limb training protocols. Future work will focus on proposing an adaptive trajectory generation algorithm with efficacy evaluation on a larger sample of subjects.

Optimal Multitrial Prediction Combination and Subject-Specific Adaptation for Minimal Training Brain Switch Designs.

Brain-Computer Interface (BCI) systems are traditionally designed by taking into account user-specific data to enable practical use. More recently, subject independent (SI) classification algorithms have been developed which bypass the subject specific adaptation and enable rapid use of the system. A brain switch is a particular BCI system where the system is required to distinguish from two separate mental tasks corresponding to the on-off commands of a switch. Such applications require a low false positive rate (FPR) while having an acceptable response time (RT) until the switch is activated. In this work, we develop a methodology that produces optimal brain switch behavior through subject specific (SS) adaptation of: a) a multitrial prediction combination model and b) an SI classification model. We propose a statistical model of combining classifier predictions that enables optimal FPR calibration through a short calibration session. We trained an SI classifier on a training synchronous dataset and tested our method on separate holdout synchronous and asynchronous brain switch experiments. Although our SI model obtained similar performance between training and holdout datasets, 86% and 85% for the synchronous and 69% and 66% for the asynchronous the between subject FPR and TPR variability was high (up to 62%). The short calibration session was then employed to alleviate that problem and provide decision thresholds that achieve when possible a target FPR=1% with good accuracy for both datasets.

A Robot-Driven Computational Model for Estimating Passive Ankle Torque With Subject-Specific Adaptation.

Robot-assisted ankle assessment could potentially be conducted using sensor-based and model-based methods. Existing ankle rehabilitation robots usually use torquemeters and multiaxis load cells for measuring joint dynamics. These measurements are accurate, but the contribution as a result of muscles and ligaments is not taken into account. Some computational ankle models have been developed to evaluate ligament strain and joint torque. These models do not include muscles and, thus, are not suitable for an overall ankle assessment in robot-assisted therapy. This study proposed a computational ankle model for use in robot-assisted therapy with three rotational degrees of freedom, 12 muscles, and seven ligaments. This model is driven by robotics, uses three independent position variables as inputs, and outputs an overall ankle assessment. Subject-specific adaptations by geometric and strength scaling were also made to allow for a universal model. This model was evaluated using published results and experimental data from 11 participants. Results show a high accuracy in the evaluation of ligament neutral length and passive joint torque. The subject-specific adaptation performance is high, with each normalized root-mean-square deviation value less than 10%. This model could be used for ankle assessment, especially in evaluating passive ankle torque, for a specific individual. The characteristic that is unique to this model is the use of three independent position variables that can be measured in real time as inputs, which makes it advantageous over other models when combined with robot-assisted therapy.

Improving Control Didactics using Multimedia

The technical evolution of university teaching has been triggered by multimedia because this new technology extends the presentation alternatives enormously. However, the didactical challenge to improve university teaching cannot be met by introducing multimedia alone but must be complemented by a thorough analysis of and by a subject specific adaptation to the didactical situation in which it is to be applied. Only if the specific didactical requirements are detected a concept for the efficient and combined use of multimedia based and traditional didactical techniques can be developed which yields the chance to make the learning process more efficient and to improve the learning results. - In this contribution several typical didactical scenarios in teaching control at the university are described and analysed with respect to didactical deficits, and possible improvements are outlined which can be obtained by modifying the traditional didactical approach and introducing multimedia based techniques.