EEG Classification with a Sequential Decision-Making Method in Motor Imagery BCI.

To develop subject-specific classifier to recognize mental states fast and reliably is an important issue in brain-computer interfaces (BCI), particularly in practical real-time applications such as wheelchair or neuroprosthetic control. In this paper, a sequential decision-making strategy is explored in conjunction with an optimal wavelet analysis for EEG classification. The subject-specific wavelet parameters based on a grid-search method were first developed to determine evidence accumulative curve for the sequential classifier. Then we proposed a new method to set the two constrained thresholds in the sequential probability ratio test (SPRT) based on the cumulative curve and a desired expected stopping time. As a result, it balanced the decision time of each class, and we term it balanced threshold SPRT (BTSPRT). The properties of the method were illustrated on 14 subjects' recordings from offline and online tests. Results showed the average maximum accuracy of the proposed method to be 83.4% and the average decision time of 2.77[Formula: see text]s, when compared with 79.2% accuracy and a decision time of 3.01[Formula: see text]s for the sequential Bayesian (SB) method. The BTSPRT method not only improves the classification accuracy and decision speed comparing with the other nonsequential or SB methods, but also provides an explicit relationship between stopping time, thresholds and error, which is important for balancing the speed-accuracy tradeoff. These results suggest that BTSPRT would be useful in explicitly adjusting the tradeoff between rapid decision-making and error-free device control.

Neuroprosthetic limb control with electrocorticography: approaches and challenges.

Advanced upper limb prosthetics, such as the Johns Hopkins Applied Physics Lab Modular Prosthetic Limb (MPL), are now available for research and preliminary clinical applications. Research attention has shifted to developing means of controlling these prostheses. Penetrating microelectrode arrays are often used in animal and human models to decode action potentials for cortical control. These arrays may suffer signal loss over the long-term and therefore should not be the only implant type investigated for chronic BMI use. Electrocorticographic (ECoG) signals from electrodes on the cortical surface may provide more stable long-term recordings. Several studies have demonstrated ECoG's potential for decoding cortical activity. As a result, clinical studies are investigating ECoG encoding of limb movement, as well as its use for interfacing with and controlling advanced prosthetic arms. This overview presents the technical state of the art in the use of ECoG in controlling prostheses. Technical limitations of the current approach and future directions are also presented.

Identifying neuron communities during a reach and grasp task using an unsupervised clustering analysis.

Recent advances in brain-machine interfaces (BMIs) have allowed for high density recordings using microelectrode arrays. However, these large datasets present a challenge in how to practically identify features of interest and discard non-task-related neurons. Thus, we apply a previously reported unsupervised clustering analysis to neural data acquired from a non-human primate as it performed a center-out reach-and-grasp task. Although neurons were recorded from multiple arrays across motor and premotor areas, neurons were found to cluster into only two groups which differ by their mean firing rate. No spatial distribution of neurons was evident in different groups, either across arrays or at different depths. Using a Kalman filter to decode arm, hand, and finger kinematics, we find that using neurons from only one of the groups resulted in higher decoding accuracy (r=0.73) than using randomly selected neurons (r=0.68). This suggests that the proposed method can be used to prune the input space and identify an optimal population of neurons for BMI tasks.

Improved BCI performance with sequential hypothesis testing.

One of the primary challenges in noninvasive brain-computer interface (BCI) control is low information transfer rate (ITR). An approach that employs a power-based sequential hypothesis testing (SHT) technique is presented for real-time detection of motor commands. Electroencephalogram (EEG) recordings obtained during a BCI task were first analyzed with a hypothesis testing (HT) method. Using serial analysis we minimized the time to determine a cued motor imagery cursor control decision. Experimental results show that the accuracy of the SHT method was above 80% for all the subjects (n = 3). The average decision time was 3.4 s, as compared with 6.0 s for the HT method. Moreover, the proposed SHT method has three times the information transfer rate (ITR) compared with the HT method.