Shouyi Wang

Using Wireless EEG Signals to Assess Memory Workload in the

Assessment of mental workload using physiological measures, especially electroencephalography (EEG) signals, is an active area. Recently, a number of wireless acquisition systems to measure EEG and other physiological signals have become available. Few studies have applied such wireless systems to assess cognitive workload and evaluate their performance. This paper presents an initial step to explore the feasibility of a popular wireless system (Emotiv EPOC headset) to assess memory workload levels in a well-known n-back task. We developed a signal processing and classification framework, which integrated an automatic artifact removal algorithm, a broad spectrum of feature extraction techniques, a personalized feature scaling method, an information-theory-based feature selection approach, and a proximal-support-vector-machine-based classification model. The experimental results show that the wirelessly collected EEG signals can be used to classify different memory workload levels for nine participants. The classification accuracies between the lowest workload level (0-back) and active workload levels (1-, 2-, 3-back) were close to 100%. The best classification accuracy for 1- versus 2-back was 80%, and 1- versus 3-back was 84%. This study indicates that the wireless acquisition system and the advanced data analytics and pattern recognition techniques are promising to achieve real-time monitoring and identification of mental workload levels for humans engaged in a wide variety of cognitive activities in the modern society.

n

There is an urgent need for a quick screening process that could help neurologists diagnose and determine whether a patient is epileptic versus simply demonstrating symptoms linked to epilepsy but actually stemming from a different illness. An inaccurate diagnosis could have fatal consequences, particularly in operating rooms and intensive care units. Electroencephalogram (EEG) has been traditionally used, as a gold standard, to diagnose patients by evaluating those brain functions that might correspond to epilepsy and other brain disorders. This research therefore focuses on developing new classification techniques for multichannel EEG recordings. Two time-series classification techniques, namely, Support Feature Machine (SFM) and Network-Based Support Vector Machine (SVM) (NSVM), are proposed in this paper to predict from EEG readings whether a person is epileptic or nonepileptic. The SFM approach is an optimization model that maximizes classification accuracy by selecting a group of electrodes (features) that has strong class separability based on time-series similarity measures and correctly classifies EEG samples in the training phase. The NSVM approach integrates a new network-based model for multidimensional time-series data with traditional SVMs to exploit both the spatial and temporal characteristics of EEG data. The proposed techniques are tested on two EEG data sets acquired from ten and five patients, respectively. Compared with other commonly used classification techniques such as SVM and decision trees, the proposed SFM and NSVM techniques provide very promising and practical results and require much less time and memory resources than traditional techniques. This study is a necessary application of data mining to advance the diagnosis and treatment of human epilepsy.

-Back Task

Epilepsy is one of the most common neurological disorders, characterized by recurrent seizures. Being able to predict impending seizures could greatly improve the lives of patients with epilepsy. In this study, we propose a new adaptive learning approach for online seizure prediction based on analysis of electroencephalogram (EEG) recordings. For each individual patient, we construct baseline patterns of normal and preseizure EEG samples, continuously monitor sliding windows of EEG recordings, and classify each window to normal or preseizure using a

Pattern- and Network-Based Classification Techniques for Multichannel Medical Data Signals to Improve Brain Diagnosis

(K)

Online Seizure Prediction Using an Adaptive Learning Approach

-nearest-neighbor (KNN) method. A new reinforcement learning algorithm is proposed to continuously update both normal and preseizure baseline patterns based on the feedback from prediction result of each window. The proposed approach was evaluated on EEG data from 10 patients with epilepsy. For each one of the 10 patients, the adaptive approach was trained using the recordings containing the first half of seizure occurrences, and tested prospectively on the subsequent recordings. Using a 150-minute prediction horizon, our approach achieved 73 percent sensitivity and 67 percent specificity on average over 10 patients. This result is shown to be far better than those of a nonupdate prediction scheme and two native prediction schemes.

Early Detection of Numerical Typing Errors Using Data Mining Techniques

This paper studies the applications of data mining techniques in early detection of numerical typing errors by human operators through a quantitative analysis of multichannel electroencephalogram (EEG) recordings. Three feature extraction techniques were developed to capture temporal, morphological, and time-frequency (wavelet) characteristics of EEG data. Two most commonly used data mining techniques, namely, linear discriminant analysis (LDA) and support vector machine (SVM), were employed to classify EEG samples associated with correct and erroneous keystrokes. The leave-one-error-pattern-out and leave-one-subject-out cross-validation methods were designed to evaluate the in- and cross-subject classification performances, respectively. For the in-subject classification, the best testing performance had a sensitivity of 62.20% and a specificity of 51.68%, which were achieved by SVM using morphological features. For the cross-subject classification, the best testing performance was achieved by LDA using temporal features, based on which it had a sensitivity of 68.72% and a specificity of 49.45%. In addition, the receiver operating characteristic (ROC) analysis revealed that the averaged values of the area under ROC curves of LDA and SVM for the in- and cross-subject classifications were both greater than 0.60 using the EEG 300 ms prior to the keystrokes. The classification results of this study indicated that the EEG patterns of erroneous keystrokes might be different from those of the correct ones. As a result, it may be possible to predict erroneous keystrokes prior to error occurrence. The classification problem addressed in this study is extremely challenging due to the very limited number of erroneous keystrokes made by each subject and the complex spatiotemporal characteristics of the EEG data. However, the outcome of this study is quite encouraging, and it is promising to develop a prospective early detection system for erroneous keystrokes based on brain-wave signals.

Online Prediction of Driver Distraction Based on Brain Activity Patterns

This paper presents a new computational framework for early detection of driver distractions (map viewing) using brain activity measured by electroencephalographic (EEG) signals. Compared with most studies in the literature, which are mainly focused on the classification of distracted and nondistracted periods, this study proposes a new framework to prospectively predict the start and end of a distraction period, defined by map viewing. The proposed prediction algorithm was tested on a data set of continuous EEG signals recorded from 24 subjects. During the EEG recordings, the subjects were asked to drive from an initial position to a destination using a city map in a simulated driving environment. The overall accuracy values for the prediction of the start and the end of map viewing were 81% and 70%, respectively. The experimental results demonstrated that the proposed algorithm can predict the start and end of map viewing with relatively high accuracy and can be generalized to individual subjects. The outcome of this study has a high potential to improve the design of future intelligent navigation systems. Prediction of the start of map viewing can be used to provide route information based on a drivers needs and consequently avoid map-viewing activities. Prediction of the end of map viewing can be used to provide warnings for potential long map-viewing durations. Further development of the proposed framework and its applications in driver-distraction predictions are also discussed.

Machine Learning Algorithms in Bipedal Robot Control

Over the past decades, machine learning techniques, such as supervised learning, reinforcement learning, and unsupervised learning, have been increasingly used in the control engineering community. Various learning algorithms have been developed to achieve autonomous operation and intelligent decision making for many complex and challenging control problems. One of such problems is bipedal walking robot control. Although still in their early stages, learning techniques have demonstrated promising potential to build adaptive control systems for bipedal robots. This paper gives a review of recent advances on the state-of-the-art learning algorithms and their applications to bipedal robot control. The effects and limitations of different learning techniques are discussed through a representative selection of examples from the literature. Guidelines for future research on learning control of bipedal robots are provided in the end.

Reinforcement Learning Control for Biped Robot Walking on Uneven Surfaces

Biped robots based on the concept of (passive) dynamic walking are far simpler than the traditional fullyI controlled walking robots, while achieving a more natural gait and consuming less energy. However, lightly actuated dynamic walking robots, which rely on the natural limit cycle of their mechanical structure, are very sensitive to ground disturbances. Already a very small step down can cause the robot to lose stability. In this paper, we investigate the use of reinforcement learning to make a dynamic walking robot more robust against ground disturbances. The learning controller is applied to a simulated two-link biped which is an abstraction of a mechanical prototype developed at the Delft Biorobotics Laboratory. The learning controller has been designed such that it can be applied as a straightforward extension of the proportionalI-derivative (PD) controller currently used to drive the robots pneumatic actuators. The learning controller is therefore suitable for the future implementation in the robot hardware. Simulation results demonstrate that the biped quickly learns to overcome step-down disturbances on the floor up to 10% of the leg length, without compromising the natural walking style provided by the PD controller, which was optimized for walking on an even surface.

A novel reinforcement learning framework for online adaptive seizure prediction

Epileptic seizure prediction is still a very challenging and unsolved problem for medical professionals. The current bottleneck of seizure prediction techniques is the lack of flexibility for different patients with an incredible variety of epileptic seizures. This study proposes a novel self-adaptation mechanism which successfully combines reinforcement learning, online monitoring and adaptive control theory for seizure prediction. The proposed method eliminates a sophisticated threshold-tuning/optimization process, and has a great potential of flexibility and adaptability to a wide range of patients with various types of seizures. The proposed prediction system was tested on five patients with epilepsy. With the best parameter settings, it achieved an averaged accuracy of 71.34%, which is considerably better than a chance model. The autonomous adaptation property of the system offers a promising path towards development of practical online seizure prediction techniques for physicians and patients.