Brain Computer Interface / BCI / Neural Control Int. / NCI / Mind Machine Int. / MMI / Direct Neural Int. / DNI / Brain Machine Int. / BMI
Fatemeh Ghomi; Amin Mahnam; Mohammad Reza Yazdchi
Volume 12, Issue 2 , September 2018, , Pages 97-109
Abstract
Over the past few decades, the brain-computer interfaces (BCI) based on motor imagery has been widely developed to help people with motor disability. The advantage of this type of BCI as an endogenous system is, no need for external stimulation, and natural control. One of the major challenges to make ...
Read More
Over the past few decades, the brain-computer interfaces (BCI) based on motor imagery has been widely developed to help people with motor disability. The advantage of this type of BCI as an endogenous system is, no need for external stimulation, and natural control. One of the major challenges to make these systems practical is to reduce the number of recording electrodes. In this study, only two EEG channels (C3 and C4) were used for detecting the imagery of left and right-hand movements. The features used were band powers (BP), some time domain parameters (TDP) and an adaptive autoregressive model (AAR). For classification, linear discriminant analysis (LDA), a well-known and simple classifier was used.The data was taken from the third BCI Competition. Our results confirm that BP features provide the most robust and effective features for accurate recognition. It was shown that combining the BP with TDP and AAR features can improve the accuracy of classification. However, implementing BP and TDP features is proposed for online classification where short computational cost is important. A maximum steepness of the mutual information (STMI) of 0.2582 was achieved in this study that could win the second place in the BCI Competition III. Left and right motor imagery (MI) tasks can be discriminated with an average classification accuracy of 85% and Kappa of 70%.
Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Alireza Mirjalili; Vahid Abootalebi; Mohammad Taghi Sadeghi
Volume 8, Issue 4 , February 2015, , Pages 305-323
Abstract
In recent years, Brain-Computer Interface (BCI) has been noted as a new means of communication between the human brain and his surroundings. In order to set up such a system, the collaboration of several blocks, such as data recording, signal processing and user interface are needed. The signal processing ...
Read More
In recent years, Brain-Computer Interface (BCI) has been noted as a new means of communication between the human brain and his surroundings. In order to set up such a system, the collaboration of several blocks, such as data recording, signal processing and user interface are needed. The signal processing block, includes two units of preprocessing and pattern recognition. Pattern recognition block itself involves two phases: feature extraction and classification. In this paper, the sparse representation based classification (SRC) has been used in the classification block. There are two important issues in using the SRC. These are creating an appropriate dictionary matrix and adopting a proper method for finding the sparse solution for an input data. In this research study, the dictionary matrix is formed by extracting an optimal set of features from the training data. Toward this goal, the common spatial patterns algorithm (CSP) is first used. Sensitivity to noise and the over learning phenomena are the main drawbacks of the CSP algorithm. In order to remove these problems, the regularized common spatial patterns algorithm (RCSP) is employed. In previous studies in within the BCI framework, the standard BP algorithm has been used to find a sparse solution. The main disadvantage of the BP algorithm is that the method is computationally expensive. To overcome this weakness, a recently proposed algorithm namely the SL0 approach is used instead. Our experimental results show that when the number of training samples is limited, the RCSP algorithm outperforms the CSP one. Using the features derived from the RCSP, the average detection rate is in average increased by a factor of 7.53%. Our classification results also show that using the SL0 algorithm, the classification process is highly speeded up as compared to the BP algorithm while an almost equivalent accuracy is achieved.
