IEEE Systems, Man and Cybernetics Magazine - July 2020 - 34

a three-layer CNN with parameters that have been experieach class k for each trial i. The classification prediction
mentally tested and validated [33]. The first layer is con-
results in choosing the class y i for which the summed
squared distance to the centroid is the smallest, i.e.,
volutional along the temporal dimension, while the
t i ii .
subsequent one is a convolutional
y i = argmin k _ c 2 _ G k, C
along the spatial dimension, i.e.,
◆◆ FBTSC: FBTSC also exploits
over EEG electrodes. The first
more spectral information than
We then propose
temporal convolution aims at optiTSC, by using a filter bank, promizing bandpass filters and the
jecting matrices C ij, bandpass
to improve these
spatial convolution seeks to optif i lt er e d i n b a nd s 4 - 8 H z ,
Riemannian
mize spatial filters. Then, signals
8-12 Hz, ..., 36-40 Hz to the tanapproaches by
are squared, a mean pooling is pergent space using (3). Then, the
formed (to compute signals band
probabilities that the vectorworking on a bank
power), and the CNN ends with a
ized upper-triangular elements
of bandpass filters
fully connected linear classificaof the projected SPD matrix
tion layer. Overall, this CNN proS ij belongs to class k is calcusuch as the ones used
cesses EEG data similarly to the
lated using standard classifor FBCSP.
FBCSP and LDA. In contrast to
f i c a t ion a lgor it h m s w it h
FBCSP, all these filters are optiprobabilistic outputs, such as
mized simultaneously, which made
LDA or LR. Here, we used LR,
it outperform the FBCSP on motor EEG signals [33].
which directly provides such probability with its softThe shallow ConvNet uses minimally preprocessed EEG
max function. Since we did so for nine frequency
signals as input, so we filtered them in 4-40 Hz.
bands, in two classes k, we ended up with nine pairs of
probabilities. From these pairs, the four most relevant
Results
are selected using mRMR on the training set. Finally,
Figure 1 summarizes the mean performance obtained on
we multiplied the probabilities associated with each
each data set. For reference, the statistical chance levels
class k, for the selected bands only, to end up with two
[41] were estimated at 50.47% for the mental workload
probabilities, using Pki = % j ! X Pkij, where Pki is the
study (1,440 trials and 22 subjects) and 52.27% for the affecprobability of trial i to be part of class k, and Pkij is the
tive state study (40 trials and 32 subjects). Note that for staprobability of a projected SPD matrix S ij, bandpass filtistical tests (ANOVA), we checked the data sphericity
tered in frequency band j, to be part of class k. The
and used a Greenhouse-Geisser (GG) correction in ANOVA
classification prediction results in choosing the class
if needed.
y i for which Pki is the highest, i.e., y i = argmax k ^Pki h .
CNNs
Briefly, a CNN is a feedforward neural network with at
least one convolutional layer. In this type of network, information flows unidirectionally from the input to the hidden
layers and finally to the output. A recent study presented a
new type of CNN dedicated to motor task classification in
EEG: the shallow ConvNet [33]. Its architecture consists of

CSP
Data Set
Emotion-Valence

Emotion-Arousal

Workload

Workload Study
The performances obtained by each algorithm on this data set
are reported in Figure 2. We performed a two-way ANOVA
with repeated measures to evaluate the performances of factor Algorithm according to factor Calibration Type (subjectspecific versus subject-independent studies). It revealed a
main effect of Algorithm [GG(1,22) = 0.517, p = 0.001], and

FBCSP FgMDM

TSC FBFgMDM

FBTSC

CNN

61.0934

46.32

Study
Subject Specific

57.5904 59.1921 58.8734

59.4658

61.0144

Subject Independent

52.5 55.2344 47.9688

49.1406

48.4375

48.75 48.0469

Subject Specific

58.2586 59.1315 60.0404

60.0404

60.3008

60.5969 40.1531

56.25

55.7813

51.6406

53.2812

Subject Specific

67.0041 68.5089 69.9429

68.4964

70.3385

68.7242 72.7296

Subject Independent

58.0465 60.0742 58.3072

58.3072

61.2989

60.1805 63.7357

Subject Independent

55.7031 55.3125

47.5

Figure 1. The mean classification accuracy for each algorithm. The best performance of each study is in green

and the worst is in red.
34	

IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE Ju ly 2020



IEEE Systems, Man and Cybernetics Magazine - July 2020

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2020

Contents
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover1
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover2
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Contents
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 2
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 3
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 4
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 5
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 6
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 7
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 8
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 9
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 10
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 11
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 12
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 13
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 14
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 15
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 16
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 17
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 18
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 19
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 20
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 21
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 22
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 23
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 24
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 25
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 26
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 27
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 28
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 29
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 30
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 31
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 32
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 33
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 34
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 35
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 36
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 37
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 38
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 39
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 40
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 41
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 42
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 43
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 44
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 45
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 46
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 47
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 48
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 49
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 50
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 51
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 52
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 53
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 54
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 55
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 56
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover3
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover4
https://www.nxtbook.com/nxtbooks/ieee/smc_202110
https://www.nxtbook.com/nxtbooks/ieee/smc_202107
https://www.nxtbook.com/nxtbooks/ieee/smc_202104
https://www.nxtbook.com/nxtbooks/ieee/smc_202101
https://www.nxtbook.com/nxtbooks/ieee/smc_202010
https://www.nxtbook.com/nxtbooks/ieee/smc_202007
https://www.nxtbook.com/nxtbooks/ieee/smc_202004
https://www.nxtbook.com/nxtbooks/ieee/smc_202001
https://www.nxtbook.com/nxtbooks/ieee/smc_201910
https://www.nxtbook.com/nxtbooks/ieee/smc_201907
https://www.nxtbook.com/nxtbooks/ieee/smc_201904
https://www.nxtbook.com/nxtbooks/ieee/smc_201901
https://www.nxtbook.com/nxtbooks/ieee/smc_201810
https://www.nxtbook.com/nxtbooks/ieee/smc_201807
https://www.nxtbook.com/nxtbooks/ieee/smc_201804
https://www.nxtbook.com/nxtbooks/ieee/smc_201801
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1017
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0717
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0417
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0117
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1016
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0716
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0416
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0116
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1015
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0715
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0415
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0115
https://www.nxtbookmedia.com