Zuoguan Wang

Cross-subject workload classification with a hierarchical Bayes model.

Most of the current EEG-based workload classifiers are subject-specific; that is, a new classifier is built and trained for each human subject. In this paper we introduce a cross-subject workload classifier based on a hierarchical Bayes model. The cross-subject classifier is trained and tested with data from a group of subjects. In our work, it was trained and tested on EEG data collected from 8 subjects as they performed the Multi-Attribute Task Battery across three levels of difficulty. The accuracy of this cross-subject classifier is stable across the three levels of workload and comparable to a benchmark subject-specific neural network classifier.

Decoding onset and direction of movements using Electrocorticographic (ECoG) signals in humans.

Communication of intent usually requires motor function. This requirement can be limiting when a person is engaged in a task, or prohibitive for some people suffering from neuromuscular disorders. Determining a persons intent, e.g., where and when to move, from brain signals rather than from muscles would have important applications in clinical or other domains. For example, detection of the onset and direction of intended movements may provide the basis for restoration of simple grasping function in people with chronic stroke, or could be used to optimize a users interaction with the surrounding environment. Detecting the onset and direction of actual movements are a first step in this direction. In this study, we demonstrate that we can detect the onset of intended movements and their direction using electrocorticographic (ECoG) signals recorded from the surface of the cortex in humans. We also demonstrate in a simulation that the information encoded in ECoG about these movements may improve performance in a targeting task. In summary, the results in this paper suggest that detection of intended movement is possible, and may serve useful functions.

Facial Feature Tracking Under Varying Facial Expressions and Face Poses Based on Restricted Boltzmann Machines

Facial feature tracking is an active area in computer vision due to its relevance to many applications. It is a nontrivial task, since faces may have varying facial expressions, poses or occlusions. In this paper, we address this problem by proposing a face shape prior model that is constructed based on the Restricted Boltzmann Machines (RBM) and their variants. Specifically, we first construct a model based on Deep Belief Networks to capture the face shape variations due to varying facial expressions for near-frontal view. To handle pose variations, the frontal face shape prior model is incorporated into a 3-way RBM model that could capture the relationship between frontal face shapes and non-frontal face shapes. Finally, we introduce methods to systematically combine the face shape prior models with image measurements of facial feature points. Experiments on benchmark databases show that with the proposed method, facial feature points can be tracked robustly and accurately even if faces have significant facial expressions and poses.

Prior Knowledge Improves Decoding of Finger Flexion from Electrocorticographic Signals

Brain–computer interfaces (BCIs) use brain signals to convey a user’s intent. Some BCI approaches begin by decoding kinematic parameters of movements from brain signals, and then proceed to using these signals, in absence of movements, to allow a user to control an output. Recent results have shown that electrocorticographic (ECoG) recordings from the surface of the brain in humans can give information about kinematic parameters (e.g., hand velocity or finger flexion). The decoding approaches in these studies usually employed classical classification/regression algorithms that derive a linear mapping between brain signals and outputs. However, they typically only incorporate little prior information about the target movement parameter. In this paper, we incorporate prior knowledge using a Bayesian decoding method, and use it to decode finger flexion from ECoG signals. Specifically, we exploit the constraints that govern finger flexion and incorporate these constraints in the construction, structure, and the probabilistic functions of the prior model of a switched non-parametric dynamic system (SNDS). Given a measurement model resulting from a traditional linear regression method, we decoded finger flexion using posterior estimation that combined the prior and measurement models. Our results show that the application of the Bayesian decoding model, which incorporates prior knowledge, improves decoding performance compared to the application of a linear regression model, which does not incorporate prior knowledge. Thus, the results presented in this paper may ultimately lead to neurally controlled hand prostheses with full fine-grained finger articulation.

Cell detection and tracking for micromanipulation vision system of cell-operation robot

The paper discusses cell detection and tracking in a microscope scene for a cell-operation robot. An approach integrating a morphological operator and Bayesian estimation technique is used for cell detection and tracking. One of the problems in the implementation is difficulty in obtaining accurate information of cells. We introduce classification criteria which rely on both shape and texture information of the cell to get fine results, and we present geometric constraints incorporating a nonlinear diffusion process to reduce the drift of points during the tracking procedure. A preliminary test is conducted on the system. Experimental results show the effectiveness of the proposed methods.

Decoding Finger Flexion from Electrocorticographic Signals Using a Sparse Gaussian Process

A brain-computer interface (BCI) creates a direct communication pathway between the brain and an external device, and can thereby restore function in people with severe motor disabilities. A core component in a BCI system is the decoding algorithm that translates brain signals into action commands of an output device. Most of current decoding algorithms are based on linear models (e.g., derived using linear regression) that may have important shortcomings. The use of nonlinear models (e.g., neural networks) could overcome some of these shortcomings, but has difficulties with high dimensional feature spaces. Here we propose another decoding algorithm that is based on the sparse gaussian process with pseudo-inputs (SPGP). As a nonparametric method, it can model more complex relationships compared to linear methods. As a kernel method, it can readily deal with high dimensional feature space. The evaluations shown in this paper demonstrate that SPGP can decode the flexion of finger movements from electrocorticographic (ECoG) signals more accurately than a previously described algorithm that used a linear model. In addition, by formulating problems in the bayesian probabilistic framework, SPGP can provide estimation of the prediction uncertainty. Furthermore, the trained SPGP offers a very effective way for identifying important features.

Workload Classification Across Subjects Using EEG

EEG data has been used to discriminate levels of mental workload when classifiers are created for each subject, but the reliability of classifiers trained on multiple subjects has yet to be investigated. Artificial neural network and naive Bayesian classifiers were trained with data from single and multiple subjects and their ability to discriminate among three difficulty conditions was tested. When trained on data from multiple subjects, both types of classifiers poorly discriminated between the three levels. However, a novel model, the naive Bayesian classifier with a hidden node, performed nearly as well as the models trained and tested on individuals.

An EEG workload classifier for multiple subjects

EEG data has been used to discriminate levels of mental workload when classifiers are created for each subject, but the reliability of classifiers trained on multiple subjects has yet to be investigated. Artificial neural network and naive Bayesian classifiers were trained with data from single and multiple subjects and their ability to discriminate among three difficulty conditions was tested. When trained on data from multiple subjects, both types of classifiers poorly discriminated between the three levels. However, a novel model, the naive Bayesian classifier with a hidden node, performed nearly as well as the models trained and tested on individuals. In addition, a hierarchical Bayes model with a higher level constraint on the hidden node can further improve its performance.