Journal of Neuroscience Methods 255 (2015) 8591

Contents lists available at ScienceDirect

Journal of Neuroscience Methods

journal homepage: www.elsevier.com/locate/jneumeth

Computational Neuroscience

Optimizing spatial patterns with sparse lter bands for

motor-imagery based braincomputer interface
Yu Zhang a, , Guoxu Zhou b , Jing Jin a , Xingyu Wang a , Andrzej Cichocki b,c
a
Key Laboratory for Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology,
Shanghai 200237, China
b
Laboratory for Advanced Brain Signal Processing, RIKEN Brain Science Institute, Wako-shi, Saitama 351-0198, Japan
c
System Research Institute, Polish Academy of Sciences, Warsaw 00-901, Poland

h i g h l i g h t s

This study proposes a sparse lter band common spatial pattern (SFBCSP) for optimizing the spatial patterns.
Experimental results on two public EEG datasets (BCI Competition III dataset IVa and BCI Competition IV IIb) conrm the effectiveness of SFBCSP.
The optimized spatial patterns by SFBCSP give overall better MI classication accuracy in comparison with several competing methods.
Our study suggests that the proposed SFBCSP is a potential method for improving the performance of MI-based BCI.

a r t i c l e

i n f o

Article history:
Received 21 May 2015
Received in revised form 3 August 2015
Accepted 5 August 2015
Available online 13 August 2015
Keywords:
Braincomputer interface (BCI)
Common spatial pattern (CSP)
Electroencephalogram (EEG)
Motor imagery (MI)
Sparse regression

a b s t r a c t
Background: Common spatial pattern (CSP) has been most popularly applied to motor-imagery (MI) feature extraction for classication in braincomputer interface (BCI) application. Successful application of
CSP depends on the lter band selection to a large degree. However, the most proper band is typically
subject-specic and can hardly be determined manually.
New method: This study proposes a sparse lter band common spatial pattern (SFBCSP) for optimizing
the spatial patterns. SFBCSP estimates CSP features on multiple signals that are ltered from raw EEG
data at a set of overlapping bands. The lter bands that result in signicant CSP features are then selected
in a supervised way by exploiting sparse regression. A support vector machine (SVM) is implemented on
the selected features for MI classication.
Results: Two public EEG datasets (BCI Competition III dataset IVa and BCI Competition IV IIb) are used to
validate the proposed SFBCSP method. Experimental results demonstrate that SFBCSP help improve the
classication performance of MI.
Comparison with existing methods: The optimized spatial patterns by SFBCSP give overall better MI classication accuracy in comparison with several competing methods.
Conclusions: The proposed SFBCSP is a potential method for improving the performance of MI-based BCI.
2015 Elsevier B.V. All rights reserved.

1. Introduction
A braincomputer interface (BCI) is an advanced communication approach that assists to establish the capabilities of
environmental control for severally disabled people (Wolpaw et al.,
2002; Gao et al., 2014; Hoffmann et al., 2008; Jin et al., 2015; Zhang
et al., 2012; Rutkowski and Mori, 2015; Chen et al., 2015). BCI can
translate a specic brain activity into computer command, thereby
building a direct connection between human brain and external

Corresponding author. Tel.: +86 21 64253581.

E-mail address: yuzhang@ecust.edu.cn (Y. Zhang).
http://dx.doi.org/10.1016/j.jneumeth.2015.08.004
0165-0270/ 2015 Elsevier B.V. All rights reserved.

device. One of the most popularly adopted brain activities is eventrelated (de)synchronization (ERD/ERS) (Pfurtscheller and Neuper,
2001; Li et al., 2013), and can be typically measured by electroencephalogram (EEG). ERD/ERS can be quantied by band-power
changes occurring when subjects do motor-imagery (MI) tasks, i.e.,
imagine their limbs (left hand, right hand and foot) (Pfurtscheller
et al., 2006; Li and Zhang, 2010; Koo et al., 2015).
So far, a large number of methods have been introduced to EEG
analysis for various applications (Li et al., 2015, 2008; Arvaneh
et al., 2013; Zhang et al., 2012, 2014; Cong et al., 2010; Jin et al.,
2013; Zhang et al., 2015b). Common spatial pattern (CSP) is a very
efcient method and has been mostly applied to MI feature extraction (Ramoser et al., 2000; Higashi et al., 2012). The variance of

86

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

Fig. 1. Illustration of the proposed sparse lter band common spatial pattern (SFBCSP) algorithm for motor-imagery classication.

Table 1
Classication errors (%) obtained by the CSP, FBCSP, DFBCSP, and SFBCSP methods, respectively, for BCI Competition III dataset IVa. For each subject, the lowest error is
marked in boldface.
Subject

CSP

FBCSP

DFBCSP

aa
al
av
aw
ay
Average
p-value

20.11 3.82
2.11 1.45
29.61 4.16
6.96 2.32
7.86 2.86
13.33 2.92
p < 0.05

9.61 2.14
2.18 1.72
27.46 6.67
2.79 1.53
5.46 2.88
9.50 2.99
p = 0.14

9.68 2.36
1.54 1.21
24.86 3.22
2.18 1.38
4.71 2.35
8.59 2.10
p = 0.27

SFBCSP
8.46
1.43
22.57
1.97
5.31
7.95

2.10
1.09
4.87
1.24
2.96
2.45

Table 2
Classication errors (%) obtained by the CSP, FBCSP, DFBCSP, and SFBCSP methods, respectively, for BCI Competition IV dataset IIb. For each subject, the lowest error is marked
in boldface.
Subject

CSP

FBCSP

DFBCSP

B0103T
B0203T
B0303T
B0403T
B0503T
B0603T
B0703T
B0803T
B0903T
Average
p-value

23.44 5.88
44.44 6.97
47.38 8.10
1.94 1.50
11.81 4.77
30.63 5.60
16.56 4.85
13.44 3.02
18.25 4.63
23.10 5.04
p < 0.01

22.50 4.61
44.06 6.05
46.25 6.95
1.13 1.13
9.56 2.42
21.94 3.44
13.50 2.90
11.25 2.78
16.12 3.39
20.70 3.74
p < 0.01

22.06 4.48
42.75 5.89
44.81 6.50
1.19 1.43
7.56 2.14
18.31 3.17
10.50 1.83
11.38 2.93
15.25 3.33
19.31 3.52
p < 0.05

band-pass ltered signals has been known to be equal to the bandpower. Since CSP nds spatial lters to maximize the variance of
the projected signal from one class while minimizing it for another
class, it provides a natural approach to effectively estimate the discriminant information of MI (Blankertz et al., 2008). However, to
guarantee the successful application of CSP to MI classication,
a pre-specied lter band is required to accurately capture the
band-power changes resulting from ERD/ERS (Ang et al., 2008).
Unfortunately, the most proper lter band is typically subjectspecic and can hardly be determined in a manual way. A poor
selection of the lter band may result in low effectiveness of CSP
(Sun et al., 2010).
Although a wide lter band (i.e., 830 Hz) was usually adopted
for CSP in MI classication, an increasing number of studies suggested that the optimization of lter band could signicantly
improve classication accuracy (Ang et al., 2008; Sun et al., 2010;
Lemm et al., 2005; Dornhege et al., 2006; Novi et al., 2007). So
far, two types of approaches have been mainly proposed to x the
problem of lter band selection. One is simultaneous optimization

SFBCSP
21.85
41.25
44.19
1.15
7.94
17.68
9.75
11.13
14.50
18.83

4.79
5.84
6.09
1.23
2.35
3.62
1.25
3.21
3.61
3.55

of spectral lters within the CSP (Lemm et al., 2005; Dornhege et al.,
2006; Higashi and Tanaka, 2013) while another one is selection of
signicant CSP features from multiple frequency bands (Ang et al.,
2008; Novi et al., 2007; Thomas et al., 2009).
By extending CSP to state space, common spatio-spectral pattern (CSSP) (Lemm et al., 2005) was proposed to optimize a simple
FIR lter by employing a temporal delay within CSP. A further
extension of CSSP is called common sparse spectral spatial pattern
(CSSSP) (Dornhege et al., 2006), which optimizes an adaptive FIR lter simultaneously with CSP. More recently, (Higashi and Tanaka,
2013) proposed to simultaneously learn multiple FIR lters and the
associated spatial weights by maximizing a cost function extended
from CSP.
On the other hand, an alternative approach is sub-band common
spatial pattern (SBCSP) (Novi et al., 2007). Instead of simultaneously
optimizing a spectral lter within CSP, SBCSP ltered EEG signals
using multiple lter bands and extracted the sub-band CSP features for classication with score fusion. Although SBCSP achieved
superior classication accuracy over both CSSP and CSSSP

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

87

Classification accuracy (%)

100

90

80

70
CSP
FBCSP

60

DFBCSP
SFBCSP

e
ag
er
Av

90

3T

3T
B0

3T

80
B0

70

3T
B0

60

3T
B0

50

3T
B0

40

3T
B0

30

3T
B0

20
B0

10

3T

ay
B0

aw

av

aa

al

50

Fig. 2. Classication accuracies of all subjects from both BCI Competition III dataset IVa and BCI Competition IV dataset IIb. Higher accuracy on average was achieved by the
proposed SFBCSP method in comparison to the CSP, FBCSP and DFBCSP methods.

raw EEG data at a set of overlapping lter bands. Sparse regression

is implemented to supervisedly select the signicant CSP features
with class labels. A support vector machine (SVM) with linear kernel is then trained on the selected features to classify MI tasks.
The proposed SFBCSP is validated on two public EEG datasets (BCI
Competition III dataset IVa and BCI Competition IV dataset IIb), and
compared with CSP, FBCSP and DFBCSP. Experimental results show
SFBCSP achieves superior classication accuracy.

86

Classification accuracy (%)

85
84
83
82
81

2. SFBCSP for MI classication

80
79

2.1. Feature extraction by CSP

78

Common spatial pattern (CSP) is an effective method for feature

extraction in discriminating two classes of data. In recent years,
CSP has become greatly popular for the extraction of EEG features
related to motor imagery. Consider two classes of EEG samples Xi,1
and Xi,2 RCP recorded from i-th trial, where C is the number of
channels, and p denotes the number of samples. Both EEG samples
are bandpass ltered within a specied frequency band. Without
loss of generality, we assume Xi,1 and Xi,2 have been centered. Spatial covariance matrix  of the class l (l = 1, 2) is then computed
by

77
76
0.05

0.1

0.15

0.2

0.25

Fig. 3. Effects of varying the regularization parameter  on the classication accuracy obtained by the SFBCSP method for subject B0103T. The parameter value
giving the highest accuracy was highlighted in a red circle. (For interpretation of the
references to color in this gure legend, the reader is referred to the web version of
this article.)

(Novi et al., 2007), it ignored the correlation among CSP features

from different lter bands. (Ang et al., 2008) introduced a lter
bank common spatial pattern (FBCSP) to select the optimal lter
bands through estimating the mutual information among CSP features at several xed lter bands. Higher classication accuracy was
achieved by FBCSP over SBCSP. Subsequently, Thomas et al. (2009)
proposed a discriminant FBCSP (DFBCSP) using Fisher ratio to select
subject-specic lter bands instead of xed ones, which enhanced
accuracy of the FBCSP.
In this study, we will focus on further improving the second type
of approaches, i.e., CSP feature selection from multiple frequency
bands. To this end, we propose a sparse lter band common spatial pattern (SFBCSP) by exploiting sparse regression for automatic
band selection. Recently, sparse regression has been increasingly
applied to BCI and shown its efciency in ERP feature extraction to
alleviate overtting (Zhang et al., 2014). In the proposed method,
CSP features are estimated on multiple signals that are ltered from

1
Xi,l XTi,l ,
Nl
Nl

l =

(1)

i=1

where Nl is the number of trials in class l. CSP is aimed at learning

linear transforms (also called spatial lters) to maximize the ratio
of transformed data variance between the two classes:
maxJ(w) =
w

wT 1 w
wT 2 w

s.t. w2 = 1,

(2)

where   2 denotes the l2 -norm, and w RC is a spatial lter. The

maximization of Rayleigh quotient J(w) can be achieved by solving
the following generalized eigenvalue problem:
1 w = 2 w.

(3)

The eigenvectors are then obtained to form the learned spatial

The projection Z of a given EEG sample X is then
lters matrix W.
given by
T

X.
Z=W

(4)

88

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

Generally, only a small number M of projected signals are used

for feature computation. With the M rst and last rows of Z, i.e.,
Zm , m {1, . . ., 2M}, the feature vector is then formed as g = [g1 , . . .,
g2M ]T with entries

var(Zm )

2M

gm = log

h=1

(5)

var(Zh )

where var(Z) denotes the variance of each row of Z.

2.2. Sparse lter band CSP
The successful application of CSP to MI classication largely
depends on the frequency band selection for bandpass ltering of
EEG (Blankertz et al., 2008). A good selection of lter bands can considerably improve the MI classication accuracy, whereas a poor
one will result in low effectiveness of CSP (Ang et al., 2008; Thomas
et al., 2009). The optimal lter band is typically subject-specic and
can hardly be determined in a manual way. To solve the problem,
this study proposes a sparse lter band common spatial pattern
(SFBCSP) for automatic selection of lter bands.
Instead of directly using a wide lter band, we perform bandpass
ltering on raw EEG by using a set of overlapping subbands (fb1 , . . .,
fbK ). These subbands are chosen from the frequency range 440
Hz with bandwidth of 4 Hz and overlapping rate of 2 Hz, i.e., fb1 =
48 Hz, fb2 = 610 Hz, . . ., fbK = 3640 Hz where K = 17. The adopted
settings of bandwidth and overlapping rate are consistent with
those in literature (Thomas et al., 2009). CSP is then implemented
on the ltered signals at each sub-band to calculate the corresponding features by (5). As a result, 2MK features are extracted from each
EEG sample. With the extracted CSP features, we can construct the
following feature set:

g1,1

.
..

...

g1,2MK

..

..
.

G=

gN,1

...

(6)

gN,2MK

where gi,j denotes the j-th feature extracted from EEG sample at
i-th trial, and N = N1 + N2 .
Here, we propose to use the following sparse regression model
(i.e, the well-known Lasso estimate (Tibshirani, 1996)) for signicant CSP feature selection
1
u = arg min Gu y22 + u1 ,
2
u

(7)

where   1 denotes the l1 -norm, y RN is a vector containing class

labels {1, 2}, u is a sparse vector to be learned,  is a positive regularization parameter for controlling the sparsity of u, and larger
 could result in more sparse u. The coordinate descent algorithm
(Friedman et al., 2010) is adopted to solve the optimization problem
in (7). The coordinate-wise update form for u is given by:
u j S

 N

i=1
(j)

(j)
gi,j (yi y i ), 

(8)

where y i =
g u is the tted value excluding the contribud=
/ j i,d d
tion from gi,j , and S(a, ) is a shrinkage-thresholding operator:

a  if a > 

0
if |a| 
sign(a)(|a| )+ =
a
+

if a < .

(9)

satisfying (7)
We can obtain the optimized sparse vector u
through repeating the update form in (9) for j = 1, 2, . . . , D, 1, 2,

. . . until convergence. The column vectors in G corresponding to

are excluded to form a optimized feature set
those zero entries in u
that is of lower dimensionality in comparison with G. The given
G
and hence the selection of
 determines the sparsity degree of u,
CSP features.
Support vector machine (SVM) is adopted and implemented on
to train the classier. For a new test samthe optimized feature set G
RCP , the corresponding subband feature vector g
R2MK
ple X
is estimated by CSP. The optimal features are selected according to
and then fed into the trained SVM for MI classithe sparse vector u
cation. A simple linear kernel is adopted for the SVM training. Fig. 1
illustrates the proposed SFBCSP algorithm for MI classication.
3. Experimental study
3.1. Public BCI Competition datasets
3.1.1. BCI Competition III dataset IVa
This dataset is recorded for ve subjects (named aa, al, av,
aw, and ay) at 118 electrodes during right hand and foot MI
tasks. For each subject, a total of 280 trials of EEG measurements
are available (half for each class of MI). The visual cue indicating the
MI task at each trial lasted for 3.5 s. The sampling rate was 100 Hz.
More details about the dataset can be found at website http://www.
bbci.de/competition/iii/.
3.1.2. BCI Competition IV dataset IIb
This dataset is recorded from nine subjects at electrodes C3, Cz
and C4 with sampling rate of 250 Hz, during right hand and left
hand MI tasks. For comparison with the results reported in literature (Thomas et al., 2009), this study only uses the third training
sessions of the dataset, i.e., B0103T, B0203T, . . ., B0903T. For
each subject, a total 160 trials of EEG measurements are available
(half for each class of MI). In each trial, the subject was indicated by
a visual cue to perform MI task for 4.5 s. See website http://www.
bbci.de/competition/iv/ for more details about the dataset.
3.2. Evaluation scheme
The effectiveness of the proposed SFBCSP method is tested on
the two above-mentioned datasets, in comparison with the CSP,
FBCSP (Ang et al., 2008) and DFBCSP methods (Thomas et al., 2009).
On each trial, a segment is extracted starting from 0.5 to 2.5 s after
the visual cue. A wide lter band 440 Hz is adopted for CSP while
17 overlapping subbands as mentioned in section 2.2 are chosen for
FBCSP, DFBCSP and SFBCSP. For each of the two datasets, a 10 10fold cross-validation is implemented to evaluate the classication
performance. After the optimal CSP features are selected, SVM is
trained to classify the MI tasks. The regularization parameters  in
SFBCSP and C in SVM are determined by cross-validation on training
data. The number of CSP lters is set to 2 (i.e., M = 1).
3.3. Experimental results
3.3.1. BCI Competition III dataset IVa
Table 1 presents classication errors derived by the CSP, FBCSP,
DFBCSP and SFBCSP methods, respectively, for the ve subjects. All
of the three modied CSP algorithms outperformed the standard
CSP. The proposed SFBCSP method further yielded lower average
errors than those of the FBCSP and DFBCSP methods. The average
error rate reductions achieved by SFBCSP were 40.36 %, 16.32 % and
7.45 % in comparison with the CSP, FBCSP and DFBCSP methods,
respectively. The SFBCSP method obtained signicant improvement on classication accuracy compared to CSP (p < 0.05).

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

89

Fig. 4. Sparse vectors and the most signicant spatial lters learned by the SFBCSP method for each of the ve subjects in BCI Competition III dataset IVa. The feature indicated
as 1, 2, . . ., 34 correspond to the lter subbands 48 Hz, 610 Hz, . . ., 3640 Hz, respectively.

B0103T

B0203T

B0303T

B0403T

B0503T

0.8

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0.2

0
5 10 15 20 25 30

0
5 10 15 20 25 30

B0603T

0
5 10 15 20 25 30

B0703T

B0803T

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0
5 10 15 20 25 30

0
5 10 15 20 25 30

Feature index

0
5 10 15 20 25 30

5 10 15 20 25 30

B0903T

0
5 10 15 20 25 30

Values of u

5 10 15 20 25 30

Fig. 5. Sparse vectors leaned by the SFBCSP method for each of the nine subjects in BCI Competition IV dataset IIb. The feature indicated as 1, 2, . . ., 34 correspond to the
lter subbands 48 Hz, 610 Hz, . . ., 3640 Hz, respectively.

3.3.2. BCI Competition IV dataset IIb

Table 2 summarizes classication errors for the nine subjects.
The SFBCSP method yielded lower classication errors for most
subjects than those of the other three methods. The classication
performance of SFBCSP was signicantly better than those of CSP
(p < 0.01), FBCSP (p < 0.01) and DFBCSP (p < 0.05). The average error
reductions achieved by the SFBCSP method were 18.48 %, 9.03 %
and 2.49 % in comparison with the CSP, FBCSP and DFBCSP methods,
respectively.
Additionally, the classication accuracies for all subjects from
both the aforementioned datasets are depicted in Fig. 2. In summary, the proposed SFBCSP method yielded higher overall accuracy
over the CSP, FBCSP and DFBCSP methods for MI classication.
4. Discussion
In the proposed SFBCSP method, the regularization parameter
 required by sparse regression played an important role in CSP
feature selection. A too large  may exclude useful features from
the model while a too small one could not eliminate redundancy

effectively. In this study, the optimal subject-specic  was determined by cross-validation on training data. Fig. 3 presents an
example to see the effects of varying  on the classication accuracy for subject B0103T. The optimal validation accuracy was
achieved at  = 0.14 that was chosen for the subsequent test procedure. A potential problem for the proposed method is that
the cross-validation procedure is usually time-consuming and is
not applicable for small sample size datasets (Lu et al., 2010).
Instead of cross-validation, Bayesian inference provides an effective
approach to automatically and quickly estimate the model parameters under the so-called evidence framework (MacKay, 1992;
Bishop, 2006). Recently, sparse Bayesian learning (Tipping, 2001)
has been increasingly applied to the automatic determination regularization in EEG analysis (Wu et al., 2011, 2014, 2014; Zhang et al.,
2015a; Zhang and Rao, 2011). In fact, Bayesian estimation of the
optimal  may further improve the efciency of SFBCSP, which will
be investigated in our future study.
Moreover, sparse regularization has been increasingly applied
to EEG analysis, such as feature reduction (Zhang et al., 2014;
Blankertz et al., 2002), channel optimization (Arvaneh et al., 2011)

90

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

FBCSP
Value of the best feature 2

0
1
2
3
4
3

2
1
0
Value of the best feature 1
DFBCSP

Value of the best feature 2

Value of the best feature 2

Value of the best feature 2

CSP
1

0
1
2
3
4
3

2
1
0
Value of the best feature 1

1
0
1
2
3
4
3

2
1
0
Value of the best feature 1
SFBCSP

2
1
0
Value of the best feature 1

1
0
1
2
3
4
3

Fig. 6. Distributions of the most signicant two features obtained by CSP, FBCSP, DFBCSP and SFBCSP, respectively, from subject B0703T.

and trial selection (Zhang et al., 2013). This study exploited sparse
regression for lter band optimization to further improving the
MI classication accuracy. With a properly determined regularization parameter , the optimal sparse vector u can be learned
from (7) to effectively select the signicant CSP features extracted
from multiple lter subbands. Fig. 4 presents the sparse vectors
and the most signicant spatial lters learned by SFBCSP for each
subject in BCI Competition III dataset IVa. Fig. 5 shows the sparse
vectors learned by SFBCSP for each subject in BCI Competition IV
dataset IIb. It can be seen that the distribution of signicant lter
bands is sparse and subject-specic. The spatial lter corresponding to the most signicant band accurately captured the features in
the sensorimotor areas. The signicant feature appeared at multiple subbands for some subjects but at an exact subband for some
other subjects (e.g., B0403T and B0903T). This indicates that an
effective optimization of lter band for CSP is necessary to improve
the MI classication accuracy. Fig. 6 depicts distributions of the
most signicant two features derived by CSP, FBCSP, DFBCSP and
SFBCSP, respectively, from subject B0703T. All of the three modied CSP algorithms provided more separable feature distributions
in comparison with the standard CSP. The highest discriminability
of features was achieved by SFBCSP.
In recent years, multiway learning algorithms have shown
their promising potentials for the collaborative optimization in the
spatial, temporal and spectral dimensions of brain signals (Cong
et al., 2013, 2015; Zhang et al., 2013, 2014; Cichocki et al., 2008;
Wu et al., 2014; Zhou et al., 2013; Zhao et al., 2015). A combination
of collaborative multiway optimization and regularization for lter
band selection could further benet to improve the effectiveness
of CSP for MI-based BCI.
5. Conclusions
In this study, we introduced a sparse lter band common spatial
pattern (SFBCSP) algorithm to automatically select the signicant
lter bands for improving MI task classication performance in
BCI. In the proposed method, CSP features were rst estimated on

multiple signals ltered from raw EEG data at a set of overlapping

lter bands. Signicant CSP features were then selected by exploiting sparse regression and were used in SVM to classify MI tasks.
Experimental studies on two public EEG datasets (BCI Competition
III dataset IVa and BCI Competition IV dataset IIb) indicated that the
proposed SFBCSP yielded higher overall classication accuracy in
comparison with several other competing methods.
Acknowledgements
This study was supported in part by the Nation Nature Science Foundation of China under Grant 61305028, Grant 91420302,
Grant 61573142, Grant 61203127, Grant 61201124, Fundamental Research Funds for the Central Universities under Grant
WH1314023, Grant WG1414005, Grant WH1414022, the Guangdong Natural Science Foundation under Grant 2014A030308009,
and JSPS KAKENHI Grant 26730125.
References
Ang K, Chin Z, Zhang H, Guan C. Filter bank common spatial pattern (FBCSP) in
braincomputer interface. In: IEEE International Joint Conference on Neural
Networks (IJCNN 2008); 2008. p. 23907.
Arvaneh M, Guan C, Ang K, Quek C. Optimizing the channel selection and classication accuracy in EEG-based BCI. IEEE Trans Biomed Eng 2011;58(6):186573.
Arvaneh M, Cuntai G, Kai K, Chai Q. Optimizing spatial lters by minimizing withinclass dissimilarities in electroencephalogram-based braincomputer interface.
IEEE Trans Neural Netw Learn Syst 2013;24(4):6109.
Bishop C. Pattern recognition and machine learning. Springer; 2006.
Blankertz B, Curio G, Mller K. Classifying single trial EEG: Toward brain computer
interfacing. In: Proc Advances in Neural Information Processing Systems (NIPS
01) 14; 2002. p. 15764.
Blankertz B, Tomioka R, Lemm S, Kawanabe M, Mller K. Optimizing spatial lters
for robust EEG single-trial analysis. IEEE Signal Process Mag 2008;25(1):4156.
Chen L, Jin J, Zhang Y, Wang X, Cichocki A. A survey of the dummy face and human
face stimuli used in BCI paradigm. J Neurosci Methods 2015;239:1827.
Cichocki A, Washizawa Y, Rutkowski T, Bakardjian H, Phan A-H, Choi S, Lee H, Zhao
Q, Zhang L, Li Y. Noninvasive BCIs: Multiway signal-processing array decompositions. IEEE Comput 2008;41(10):3442.
Cong F, Kalyakin I, Huttunen-Scott T, Li H, Lyytinen H, Ristaniemi T. Single-trial
based independent component analysis on mismatch negativity in children. Int
J Neural Syst 2010;20(4):27992.

Y. Zhang et al. / Journal of Neuroscience Methods 255 (2015) 8591

Cong F, Phan A-H, Astikainen P, Zhao Q, Wu Q, Hietanen J, Ristaniemi T, Cichocki
A. Multi-domain feature extraction for small event-related ptentials through
nonnegative multi-way array decomposition from low dense array EEG. Int J
Neural Syst 2013;23(2):1350006.
Cong F, Lin Q, Kuang L, Gong X, Astikainen P, Ristaniemi T. Tensor decomposition of
EEG signals: a brief review. J Neurosci Meth 2015;248:5969.
Dornhege G, Blankertz B, Krauledat M, Losch F, Curio G, Mller K. Combined
optimization of spatial and temporal lters for improving braincomputer interfacing. IEEE Trans Biomed Eng 2006;53(11):227481.
Friedman J, Hastie T, Tibshirani R. Regularization paths for generalized linear models
via coordinate descent. J Stat Softw 2010;33(1):122.
Gao S, Wang Y, Gao X, Hong B. Visual and auditory braincomputer interfaces. IEEE
Trans Biomed Eng 2014;61(5):143647.
Higashi H, Tanaka T. Simultaneous design of FIR lter banks and spatial patterns for
EEG signal classication. IEEE Trans Biomed Eng 2013;60(4):110010.
Higashi H, Cichocki A, Tanaka T. Regularization using geometric information
between sensors capturing features from brain signals. In: IEEE International
Conference on Acoustics, Speech and Signal Processing (ICASSP 2012); 2012. p.
7214.
Hoffmann U, Vesin J, Ebrahimi T, Diserens K. An efcient P300-based
braincomputer interface for disabled subjects. J Neurosci Methods
2008;167(1):11525.
Jin J, Sellers EW, Zhang Y, Daly I, Wang X, Cichocki A. Whether generic model works
for rapid ERP-based BCI calibration. J Neurosci Methods 2013;212(1):949.
Jin J, Sellers E, Zhou S, Zhang Y, Wang X, Cichocki A. A P300 brain computer interface
based on a modication of the mismatch negativity paradigm. Int J Neural Syst
2015;25(3):1550011.
Koo B, Lee H, Nam Y, Kang H, Koh C, Shin H, Choi S. A hybrid NIRS-EEG system for selfpaced brain computer interface with online motor imagery. J Neurosci Methods
2015;244:2632.
Lemm S, Blankertz B, Curio G, Mller K. Spatio-spectral lters for improving the
classication of single trial EEG. IEEE Trans Biomed Eng 2005;52(9):15418.
Li J, Zhang L. Bilateral adaptation and neurofeedback for brain computer interfacesystem. J Neurosci Methods 2010;193(2):3739.
Li Y, Guan C, Li H, Chin Z. A self-training semi-supervised SVM algorithm and its
application in an EEG-based brain computer interface speller system. Pattern
Recognit Lett 2008;29(9):128594.
Li J, Liang J, Zhao Q, Li J, Hong K, Zhang L. Design of assistive wheelchair system
directly steered by human thoughts. Int J Neural Syst 2013;23(3):1350013.
Li J, Struzik Z, Zhang L, Cichocki A. Feature learning from incomplete EEG with
denoising autoencoder. Neurocomputing 2015;165:2331.
Lu H, Eng H, Guan C, Plataniotis K, Venetsanopoulos A. Regularized common spatial
pattern with aggregation for EEG classication in small-sample setting. IEEE
Trans Biomed Eng 2010;57(12):293646.
MacKay D. Bayesian interpolation. Neural Comput 1992;4(3):41547.
Novi Q, Guan C, Dat TH, Xue P. Sub band common spatial pattern for braincomputer
interface. In: Proceedings of 3rd International Conference Neural Engineering.
EEE Engineering in Medicine and Biology Society. (EMBS); 2007. p. 2047.
Pfurtscheller G, Neuper C. Motor imagery and direct braincomputer communication. Proc IEEE 2001;89(7):112334.
Pfurtscheller G, Brunner C, Schlgl A, Lopes da Silva F. Murhythm (de), synchronization and EEG single-trial classication of different motor imagery tasks.
NeuroImage 2006;31(1):1539.
Ramoser H, Muller-Gerking J, Pfurtscheller G. Optimal spatial ltering of single trial
EEG during imagined hand movement. IEEE Trans Rehab Eng 2000;8(4):4416.
Rutkowski TM, Mori H. Tactile and bone-conduction auditory brain computer interface for vision and hearing impaired users. J Neurosci Methods 2015;244:4551.

91

Sun G, Hu J, Wu G. A novel frequency band selection method for common spatial

pattern in motor imagery based brain computer interface. In: IEEE International
Joint Conference on Neural Networks (IJCNN 2010); 2010. p. 16.
Thomas K, Guan C, Lau C, Vinod A, Ang K. A new discriminative common spatial pattern method for motor imagery braincomputer interfaces. IEEE Trans Biomed
Eng 2009;56(11):27303.
Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc B
1996;58(1):26788.
Tipping M. Sparse Bayesian learning and the relevance vector machine. J Machine
Learn Res 2001;1:21144.
Wolpaw JR, Birbaumer N, McFarland DJ, Pfutscheller G, Vaughan TM.
Braincomputer interfaces for communication and control. Clin Neurophysiol
2002;113(6):76791.
Wu W, Chen Z, Gao S, Brown E. A hierarchical bayesian approach for learning sparse spatio-temporal decompositions of multichannel EEG. NeuroImage
2011;56(4):192945.
Wu W, Wu C, Gao S, Liu B, Li Y, Gao X. Bayesian estimation of ERP components from
multicondition and multichannel EEG. NeuroImage 2014;88:31939.
Wu W, Chen Z, Gao X, Li Y, Brown E, Gao S. Probabilistic common spatial patterns for multichannel EEG analysis. IEEE Trans Pattern Anal Mach Intell
2014;37(3):63953.
Wu Q, Zhang L, Cichocki A. Multifactor sparse feature extraction using convolutive
nonnegative tucker decomposition. Neurocomputing 2014;129:1724.
Zhang Z, Rao B. Sparse signal recovery with temporally correlated source vectors using sparse Bayesian learning. IEEE J Sel Top Signal Process 2011;5(5):
91226.
Zhang Y, Zhao Q, Jin J, Wang X, Cichocki A. A novel BCI based on ERP components sensitive to congural processing of human faces. J Neural Eng 2012;9(2):026018.
Zhang Y, Jin J, Qing X, Wang B, Wang X. LASSO based stimulus frequency recognition
model for SSVEP BCIs. Biomed Signal Process Control 2012;7(4):10411.
Zhang Y, Zhou G, Jin J, Wang M, Wang X, Cichocki A. L1-regularized multiway canonical correlation analysis for SSVEP-based BCI. IEEE Trans Neural Syst Rehabil Eng
2013;21(6):88796.
Zhang Y, Zhou G, Zhao Q, Jin J, Wang X, Cichocki A. Spatial-temporal discriminant
analysis for ERP-based braincomputer interface. IEEE Trans Neural Syst Rehabil
Eng 2013;21(2):23343.
Zhang Y, Zhou G, Jin J, Zhao Q, Wang X, Cichocki A. Aggregation of sparse linear discriminant analysis for event-related potential classication in braincomputer
interface. Int J Neural Syst 2014;24(1):1450003.
Zhang Y, Zhou G, Jin J, Wang X, Cichocki A. Frequency recognition in SSVEPbased BCI using multiset canonical correlation analysis. Int J Neural Syst
2014;24(4):1450013.
Zhang Y, Xu P, Cheng K, Yao D. Multivariate synchronization index for frequency
recognition of SSVEP-based braincomputer interface. J Neurosci Methods
2014;221:3240.
Y. Zhang, G. Zhou, J. Jin, Q. Zhao, X. Wang and A. Cichocki Sparse Bayesian classication of EEG for braincomputer interface, IEEE Trans Neural Netw Learn Syst
(2015), In press.
Zhang Y, Zhou G, Jin J, Wang X, Cichocki A. SSVEP recognition using common feature
analysis in braincomputer interface. J Neurosci Methods 2015b;244:815.
Zhao Q, Zhou G, Zhang L, Cichocki A, Amari SI. Bayesian robust tensor factorization for incomplete multiway data. IEEE Trans Neural Netw Learn Syst 2015.,
http://dx.doi.org/10.1109/TNNLS.2015.2423694, in press.
Zhou G, Cichocki A, Xie S. Accelerated canonical polyadic decomposition by
using mode reduction. IEEE Trans Neural Netw Learn Syst 2013;24(12):
205162.