Professional Documents
Culture Documents
Tongue-Rudder: A Glossokinetic-Potential-Based Tongue-Machine Interface
Tongue-Rudder: A Glossokinetic-Potential-Based Tongue-Machine Interface
Tongue-Rudder: A Glossokinetic-Potential-Based Tongue-Machine Interface
1, JANUARY 2012
Tongue-Rudder: A Glossokinetic-Potential-Based
Tongue–Machine Interface
Yunjun Nam, Qibin Zhao, Andrzej Cichocki, Member, IEEE, and Seungjin Choi*, Member, IEEE
Abstract—Glossokinetic potentials (GKPs) are electric potential is regarded as a suitable organ for manipulating assistive de-
responses generated by tongue movement. In this study, we use vices involving motor control. These reasons are as follows:
these GKPs to automatically detect and estimate tongue positions, 1) the tongue is directly connected to the brain by cranial nerves
and develop a tongue–machine interface. We show that a specific
configuration of electrode placement yields discriminative GKPs and the distance from the brain is relatively short; 2) the tongue
that vary depending on the direction of the tongue. We develop a lin- generally escapes severe damage in spinal cord injuries, and
ear model to determine the direction of tongue from GKPs, where in many cases, it is more slowly affected than limbs of per-
we seek linear features that are robust to a baseline drift prob- sons suffering from most neuromuscular degenerative disorder;
lem by maximizing the ratio of intertask covariance to intersession 3) the tongue consists of special muscles suitable for complex
covariance. We apply our method to the task of wheelchair con-
trol, developing a tongue–machine interface for wheelchair control, vocalization and ingestion tasks so that it can move very quickly
referred to as tongue-rudder. A teeth clenching detection system, and accurately with little fatigue; and 4) tongue movements can
using electromyography, was also implemented in the system in be hidden by the mouth cavity, which is a cosmetic advantage.
order to assign teeth clenching as the stop command. Experiments Several tongue-operated assistive devices, which benefitted
on off-line cursor control and online wheelchair control confirm from the aforementioned advantages, have been developed. The
the unique advantages of our method, such as: 1) noninvasiveness,
2) fine controllability, and 3) ability to integrate with other EEG- “tongue drive” system uses a tongue-mounted permanent mag-
based interface systems. net and magnetic-sensors-implanted dental retainer for tracking
tongue movement [2]. The inductive tongue computer interface
Index Terms—Electric wheelchair control, glossokinetic poten-
tials (GKPs), tongue–machine interface. (ITCI) observes changes in the inductance from coils (attached
to the palatal plate), which are caused by the movement of
ferromagnetic material (attached to the tongue) [3]. The “think-
a-move” system can recognize commands by analyzing acoustic
I. INTRODUCTION
patterns generated by specific tongue motions such as flicking
SSISTIVE technologies have been developed for persons of the tongue to the left/center/right gum line [4]. The “tongue-
A with limb motor disabilities and even for those with com-
plete quadriplegia, in order to provide an alternative communi-
mouse” uses piezoelectric ceramic materials, which produce
small charges when pressed, to sense a push by the tongue [5].
cation channel and help the persons in performing daily tasks The “Tonguepoint” is a special mouthpiece, and it has a small
by utilizing biosignals measured from various sensors placed on pressure-sensitive joystick on the hard palate region [6].
the persons. For several reasons, highlighted in [1], the tongue Most of the existing methods for tongue–machine interfaces
require that sensors be placed on the tongue or inside the mouth
in order to acquire signals involving tongue movement. They
are limited to only a few commands so that fine control of
Manuscript received December 28, 2010; revised April 22, 2011 and August device is not possible. In this paper, we present a novel tongue–
10, 2011; accepted October 4, 2011. Date of publication October 28, 2011; date
of current version December 21, 2011. This work was supported in part by the machine interface where we detect tongue positions solely by us-
National Research Foundation (NRF) of Korea under Grant 2011-0018283 and ing glossokinetic potentials (GKPs) that are EEG signals related
Grant 2011-0018284, by the NRF World Class University Program under Grant to tongue movements. In contrast to existing tongue–computer
R31-10100, and by the TÜBİTAK under Project 110E232. A portion of this
work was carried out while Y. Nam was visiting the Advanced Brain Signal interface systems, our method does not require any extra sensors
Processing Lab, Brain Science Institute, RIKEN, Wako-shi, Japan. Asterisk in- placed inside mouth, which might be more comfortable to ac-
dicates corresponding author. quire signals involving tongue movements. Moreover, as will be
Y. Nam is with the School of Interdisciplinary Bioscience and Bioengineer-
ing, Pohang University of Science and Technology, Pohang 790-784, Korea shown, our method enables fine-grained device control with an
(e-mail: druid@postech.ac.kr). accuracy (in terms of difference between the estimated direction
Q. Zhao is with the Advance Brain Signal Processing Lab, Brain and the true direction) of approximately 20◦ .
Science Institute, RIKEN, Wako-shi, Saitama 351-0198, Japan (e-mail:
qbzhao@brain.riken.jp). In our preliminary results reported in [7], we showed that a
A. Cichocki is with the Advance Brain Signal Processing Lab, Brain Sci- specific configuration of electrode placement yields discrimi-
ence Institute, RIKEN, Saitama 351-0198, Japan, and also with the Systems native GKPs that vary in accordance with the direction of the
Research Institute, Polish Academy of Science, 01-447 Warszawa, Poland
(e-mail: a.cichocki@brain.riken.jp). tongue. Thus, GKP is a direct continuous interpretation of the
∗ S. Choi is with the Department of Computer Science and Division of
direction of the tongue, leading to fine-grained device control.
IT Convergence Engineering, Pohang University of Science and Technology, In this paper, we develop a linear model to associate GKP with
Pohang 790-784, Korea (e-mail: seungjin@postech.ac.kr).
Color versions of one or more of the figures in this paper are available online the direction of the tongue, where we seek linear features that
at http://ieeexplore.ieee.org. are robust to the baseline drift (the gradual and linear shift in
Digital Object Identifier 10.1109/TBME.2011.2174058
affects the potential levels on the scalp (see [8, pp. 114–117]).
Vanhatalo et al. [10] also reported that GKP can be removed
by insulating the surface of the tongue in order to block the
discharge. This GKP has been studied primarily within EEG
analysis. Since it originates from a noncerebral region, it can
interfere with the observation of brain activities. Thus, it has
been ignored or actively removed during the recording or the
preprocessing stages, likewise other artifacts such as EMG or
electrooculogram (EOG) [11].
GKP has attracted little attention, compared to EMG or EOG,
since GKP is easily regulated than EMG or EOG. First, GKP
can be consciously suppressed by the subject. Unlike EOG ar-
tifacts, which are caused by an involuntary eye–ball movement
or blinking of the eye, GKP is caused by a voluntary tongue
movement. Thus, it can easily be regulated unless the subject
has special symptoms such as Parkinson’s disease or tremor
disorder. Second, GKP can easily be removed by simple high-
pass filtering. EMG is also a consciously regulatable artifact
Fig. 1. Overview of the “tongue-rudder” system. GKP is analyzed to estimate
the direction of the tongue (in the range between − π2 and π2 ) that facilitates like GKP. However, once it is generated, it severely hampers
smooth control of the wheelchair. The electromyograph (EMG) signals that EEG analysis because it contaminates all frequency bands in
arise when clenched are used for sending special commands such as “stop” or the millivolt scale [12]. In contrast, GKP has influence only
“turn ON/OFF.”
on low-frequency bands, so it induces less deterioration and can
easily be removed by simple high-pass filtering with a cutoff fre-
the baseline over a specified period of time), by maximizing quency at 4 Hz. Because of these reasons, GKP has been studied
the ratio of intertask covariance to intersession covariance. We less and considered only when the EEG analysis is related to
apply our method to the task of wheelchair control, developing vocalization or ingestion tasks [13], [14].
a tongue–machine interface for wheelchair control, referred to
as tongue-rudder.
The rest of this paper is organized as follows. In Section II, we B. Seesaw-Like GKPs
first briefly explain GKP that is a slow wave response caused by During our GKP experiment, we observed the correlation
tongue movement. We then describe seesaw-like GKPs that vary between GKP and horizontal tongue movements [see Fig. 2(a)],
in accordance with tongue positions that are observed under a with the specific electrode placement, as depicted in Fig. 2(b). In
specific configuration of electrode placement. These seesaw-like other words, we observed that when the tongue touches the lips
GKPs are one of the main topics discussed in this paper, because and then moves in a horizontal direction, the dc levels of EEG
they enable us to develop an EEG-based tongue–machine inter- signals recorded from two earlobes move to opposite directions
face. In Section III, we develop a method for interpreting GKP and their magnitude is linearly proportional to the direction of
in terms of the direction of the tongue. We describe a linear the tongue, as can be seen in Fig. 4.
feature extraction method that is robust to the baseline drift. To validate the seesaw-like patterns of this GKP, we per-
In Section IV, we describe the implementation details of our formed two experiments. In the first experiment, we investigated
tongue-rudder system, an overview of which is graphically de- the spatial pattern of GKP, while in the second experiment, we
picted in Fig. 1. The tongue-based interface is similar in function clarified the continuous relationship between the direction of
to a steering wheel and allows the wheelchair to turn smoothly. the tongue and GKP.
The clenching-based interface is similar to a brake as it toggles For the experiments, EEG signals were recorded using a
between run and stop states. Section V presents experiments on g.USBamp (g.tec, Graz, Austria) device system with Ag/AgCl
off-line cursor control and online wheelchair control. Finally, electrodes. Channel position has already been indicated in
conclusions are drawn in Section VI. Fig. 2(b). This specific layout of electrodes was applied for
maximizing the sensitivity to antisymmetric potential changes.
II. GKPS To remove oscillations of higher frequency ranges, signals were
low-pass filtered with a cutoff frequency at 4 Hz by a Butter-
In this section, we briefly explain GKPs, and then, introduce
worth filter.
seesaw-like GKPs that are used in our system.
In the first experiment, three cues assigned to the left (L,
touching inside of the left cheek), front (F, touching lips on the
A. Background: GKPs
front side), and right (R, touching inside of the right cheek)
GKPs are electric potential response generated by tongue were sequentially shown to the subjects. These points (L, F, and
movement [8], [9]. The tip of the tongue has a negative electric R) will be used as the referential positions in the entire paper.
charge with respect to the root. As a result, if the tongue touches In accordance with the cues, the subjects continuously moved
the palate or inside the mouth, a discharge is generated that their tongue from L to F to R; this action is similar to that of
292 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 59, NO. 1, JANUARY 2012
difference between the two channels [see Fig. 4(d)] was almost
linearly proportional to the direction of the tongue.
In these two experiments, we observed that GKP evoked by
horizontal tongue movements follows a seesaw-like behavior
such that when one side goes up the other side goes down, and
the lifted heights are proportional to the distance from the center.
Thus, we refer to this pattern as seesaw-like GKPs.
The origin of this seesaw-like phenomenon can be explained
by the electric field generated by the charge on the tongue.
The tip of the tongue has a negative charge with respect to the
root. As a result, when the tongue touches the inside of the
Fig. 2. (a) If the tongue touches the inside of the cheek (buccal wall) by passing
cheek, the charge decreases the potential on the skin near the
in between the upper and the lower teeth, a potential alteration is generated. We contact point. In addition, its effect is maximum at the con-
analyzed this potential to detect the direction of the tongue within the range tact point and decreases with the distance from the point (see
−π/2 (touching the inside of the left cheek) to π/2 (touching the inside of the
right cheek). (b) Channel position for EEG recording during the implementation
[16, p. 294]). Thus, if the negatively charged tongue moves to the
step. The reference electrode was mounted on top of the head (Cz, dashed circle), left side, the potentials on the electrodes in the left hemisphere
and two ground electrodes were mounted on the forehead and at the back of will gradually decrease and the potentials on the electrodes in
the head (POz, AFz, black dashed circle). Then, signals were recorded from
temporal regions (F7/8, T7/8, P7/8, and left/right earlobes (LE/RE), and black
the right hemisphere will increase proportionally, and vice versa,
circle) for measuring antisymmetric GKP. (a) Tongue movements during the as shown in Fig. 4.
experiments. (b) Channel location. Moreover, the pattern of the seesaw-like GKP was clearly
distinguishable from known EEG phenomena such as event-
related potential (ERP) and slow cortical potential (SCP). This
seesaw-like GKP has persistent dc potential shift, unlike various
licking the lips. They were also instructed not to move their chin rhythmic EEG activities (e.g., δ ∼ γ rhythms or a sleep spin-
during the tongue movements. This was done in order to prevent dle). In addition, the scale of GKP had a range of several tens
shifting or tilting of the electrodes. The length of each cue was of microvolts. In the results of the first experiment, the maxi-
4 s and the single session consisting of three different cues was mum voltage difference between L and R for each subject was
repeated eight times. 82.1 μV (F7 for subject 1), 25.8 μV (T8 for subject 2), and
In each session, we measured the potential differences while 41.2 μV (T7 for subject 3). By way of comparison, ERP ex-
the tongue moved from L to F and from F to R, and their mean hibits amplitudes ranging only between 2 and 20 μV (see [17, p.
differences on each channel were visualized in Fig. 3 using the 639 and p. 643]). The antisymmetric spatial patterns represented
EEGLAB topoplot function [15]. In both the cases of L to F in Fig. 3 are the final distinctive feature of the seesaw-like GKP.
and F to R, the potential levels on the electrodes of the left
hemisphere increased, while the levels on the electrodes of right
hemisphere decreased. In addition, the absolute value of the III. METHOD
potential changes on the electrodes increased toward the lateral
direction. The potentials of the electrodes on the LE and the RE, A. Feature Extraction
which are located closely to the cheek and the tongue, showed As mentioned in the previous section, potential differences
larger differences. between the electrodes in the left and right hemispheres are
The second experiment was carried out to show the con- linearly proportional to the direction of the tongue. These po-
tinuous relationship between the direction of the tongue and tential differences can be used to detect the tongue’s position. As
potential changes [7]. The subject was asked to move his tongue shown in Fig. 4, simple subtraction between two channels could
to the direction where the cue was located while maintaining a be an acceptable choice, but for maximizing the discriminabil-
contact between the tongue and the inside of the cheek. At the ity, we applied a feature extraction approach with multichannel
beginning of each trial, the cue appeared from the right side and recording.
moved to the left side along a semicircular path. Then, the cue Fig. 5(a) shows EEG signals recorded during eight sessions
was returned to the right side by the same path, as shown in of tongue movements. A single session consisted of four kinds
Fig. 4(a). of tasks, namely: relax, left (L), front (F), and right (R). During
Signals recorded on the LE and the RE are shown in Fig. 4(b) L/F/R tasks, the tongue touched each position for 4 s. During
and (c). We selected these channels because they showed the the relax state, the tongue moved to the center of the mouth
largest potential changes in the first experiment. We denoted while remaining not in contact with any place inside the cheek
the signal from the LE by x1 (t) and the signal from the RE or the palate. Our objective is to obtain an 1-D feature that is
by x2 (t). When the tongue was moving from the right side to well-matched to tongue positions. In the case of the signals in
the left side (0–6 s), the signal of x1 (t) continuously decreased, Fig. 5(a), the corresponding feature will have the shape of a
while x2 (t) increased in similar continuous manner. On the ladder with three steps (each step corresponding to L, F, and
other hand, when the tongue was moving to the right side (8– R). This ladder shape will be repeated eight times for eight
14 s), x1 (t) increased while x2 (t) decreased. The potential sessions.
NAM et al.: TONGUE-RUDDER: A GLOSSOKINETIC-POTENTIAL-BASED TONGUE–MACHINE INTERFACE 293
Fig. 3. Potential differences for different positions of the tongue. Potential differences while the tongue moves from L to F are presented on the left side and
differences from F to R are presented on the right side by topographical plots. For enhancing the visibility, we applied different scale ranges for each plot. It can
be seen that the antisymmetric spatial patterns are common, but their intensities are varied. (a) Subject 1. (b) Subject 2. (c) Subject 3.
Fig. 5. (a) Signals from eight EEG channels on temporal regions are recorded and low-pass filtered below 8 Hz. Colored boxes below the plot show the labels
for the signals (black: relax, red: left, green: front, and blue: right). The signals were recorded during eight sessions. Within each session, signals from the left
side (LE, F7, T7, P7) and the right side (RE, F8, T8, P8) show antisymmetric potential change when the tongue is moving to the right. Each signal is showing
drift of baselines, especially for P7 and P8. (b) Projection result from the signal of (a) to a feature vector from ordinary PCA is plotted in blue. The green line is
the expected result when the projection results for the first session are maintained without drift. These expected results are set to μ L (1 ) , μ F (1 ) , and μ R (1 ) , the
feature’s mean values for the first session. The absolute error between the two lines is plotted in red. Note that projection results are slowly biasing to the negative
direction and errors are increasing. (c) Signals from LE and RE from a different experiment were plotted in 2-D scatter plot for explaining the patterns for GKP and
the drift. Each “x” mark is colored by a corresponding task (black: relax, red: left, green: front, and blue: right. Routes for session 1 are drawn by small arrows).
When the tongue moves from left (red) to front (green) and right (blue), corresponding points move in the direction of wT . This pattern is maintained during each
of the three sessions, but due to the drift, points gradually move in the direction of wS . We tried to obtain a novel feature vector that is sensitive to changes for
the direction of wT , while insensitive to changes for wS , by using simultaneous diagonalization. (d) A projection result using the novel feature vector is plotted.
Compared to the result of (c), it can be seen that the drift is reduced and the error has also been decreased.
(wT ). This direction can also be extracted by the previous PCA wT ΣS w. Therefore, a new projection vector should be able to
approach. If there is no drift, regions for each task should remain maximize wT ΣT w, while minimizing wT ΣS w.
at the same positions, illustrated by the boundary for session 1. The intertask covariance matrix reflects the distribution of
However, because of the drift, while the “SESSION” continues data points within each session and it can be calculated by the
from 1 to 2 to 3, the points slowly drift to the lower left direction sum of the covariance matrices from each session as follows:
(wS ).
If the new feature obtained by the projection w T X is sensitive
S
corner of the mouth (where the upper lip meets the lower lip,
the detection result may correspond to −π/4), the wheelchair
steered to the left, analogous to turning the steering wheel to the
left in a moving car. Due to this fine-grained controllability, the
subjects could drive the wheelchair smoothly, even on curved
pathways, as will be shown in Section V.
TABLE I
PERFORMANCE RESULT FOR CUE FOLLOWING EXPERIMENT
Fig. 10. Real-time online experiment. (a) We marked a track having the same topology as the Long Beach circuit map on the ground. (b) Implemented tongue-
rudder system. Each subject was asked to maneuver the wheelchair that followed the track. Time records for checkpoints are presented in (c), while recorded
signals and analysis result are presented in (d). The top figure in (d) is actually recorded signals during the driving, and the middle figure is the predicted direction
of the tongue. If the tongue moves to the left side, it shows higher values and the wheelchair turns left faster, and vice versa. In the bottom figure, the sum of
variances for eight channels was plotted. The higher value means stronger teeth clenching. The driving began [(S) ∼ (1)] and finished [(9) ∼ (E)] with 2 s of teeth
clenching. When passing through the left turning curves [at (2), (5), and (7), marked with flags in (c)], the subject moved the tongue to the left side, then the
system successfully recognized it, shown as the higher values in the middle figure of (d). (a) The marked track for the experiment. (b) An actually implemented
wheelchair system. (c) Time records for checkpoints. (d) Recorded signal and analysis result.
We marked the same topology as the Long Beach circuit line cursor control and online wheelchair control confirmed the
map on the ground, then asked the subjects to follow the track. following advantages of our method.
The subjects were able to control and maneuver the wheelchair Tongue-rudder can detect the tongue’s position from outside
successfully. A video clip of this experiment is available the mouth. Many tongue-related interface systems inject sensors
at http://mlg.postech.ac.kr/research/tongue.html. The recorded or magnets into the tongue or teeth, but injecting devices into
EEG signals and analysis results obtained while one subject was parts of the mouth can cause discomfort, and it may not be
manipulating the wheelchair are shown in Fig. 10(d). hygienic. However, this interface only uses EEG signals that are
recorded from the scalp.
The developed interface can precisely distinguish between
VI. CONCLUSION different directions of the tongue. Conventional tongue-based
In this study, we identified the seesaw-like GKP that revealed interfaces generate discretely separated commands (similar to
antisymmetric potential changes triggered by contact between toggling switches), but our interface enables continuous analog
the tongue and the inside of a cheek. We showed that the anti- manipulation. This feature has been smoothly integrated with a
symmetric behavior of GKP is linearly proportional to the di- wheelchair system and enables the wheelchair to be effortlessly
rection of the tongue under a specific configuration of electrode maneuvered, in a manner similar to driving with a steering
placement. wheel.
By using this novel potential response, we developed a Another expected advantage is that the system can easily
tongue–machine interface for wheelchair control, referred to be integrated with EEG-based brain–computer interface sys-
as tongue-rudder. We also developed a linear detection model, tems. The only major physical difference between our system
where we sought linear features that are robust to a baseline and conventional EEG systems lies in the positioning of the
drift problem by maximizing the ratio of intertask covariance ground/reference electrodes. Furthermore, the system uses a
to intersession covariance. The EMG detection system was then low-frequency band for the analysis, so the interference with
integrated into the system to facilitate the recognition of three ordinary EEG-based BCI Systems, such as motor imagery or
essential commands for wheelchair control. Experiments on off- P300, could be negligible. The integrated approach would be
NAM et al.: TONGUE-RUDDER: A GLOSSOKINETIC-POTENTIAL-BASED TONGUE–MACHINE INTERFACE 299
beneficial for extending the recognition capacity of the inter- [18] P. Tallgren, “Dc-EEG for routine clinical use: Methods and clinical im-
face if the target subject still has motility of the tongue. pact,” Ph.D. dissertation, Helsinki Univ. of Technol., Helsinki, Finland,
Dec. 2006.
In spite of aforementioned advantages, there are a number [19] P. Tallgren, S. Vanhatalo, K. Kaila, and J. Voipio, “Evaluation of commer-
of challenges that remain to be overcome in order to realize a cially available electrodes and gels for recording of slow EEG potentials,”
practical and reliable interface. First of all, the feature that we Clin. Neurophysiol., vol. 116, no. 4, pp. 799–806, 2005.
[20] T. Yagi, Y. Kuno, K. Koga, and T. Mukai, “Drifting and blinking compen-
suggested to reduce the accumulation of drift cannot achieve sation in electro-oculography (EOG) eye-gaze interface,” in Proc. IEEE
complete elimination. The most promising approach is to sep- Int. Conf. Syst., Man, Cybern., Taipei, Taiwan, 2006, pp. 3222–3226.
arate the signals into glossokinetic component and drift com- [21] K. Choi and A. Cichocki, “Control of a wheelchair by motor imagery
in real time,” in Proc. 9th Int. Conf. Intelligent Data Eng. Automated
ponent, then use the drift component to update the baselines. Learning (IDEAL), Daejeon, Korea, 2008, pp. 330–337.
Second, the combination of the tongue-based interface and the
clenching-based interface is incomplete. Test subject 4 could not
use the combination method because his clenching task causes
fluctuation of the baselines. This implies that individual anatom- Yunjun Nam received the B.Sc. degree in electrical
and computer engineering from Hanyang University,
ical differences can result in inability to operate the interface. In Seoul, Korea, in 2008. He is currently working toward
this section, we feel that further experiments with more subjects, the Ph.D. degree at the School of Interdisciplinary
Bioscience and Bioengineering, Pohang University
including paralyzed patients, are required before this interface of Science and Technology, Pohang, Korea.
can be implemented for general use. His current research interests include brain–
computer interface and electroencephalography
REFERENCES signal processing.