A technique to i n v e s t i g a t e the three-dimensional kinesiology

of the human temporomandibular joint

S. S i e g l e r , PhD.,* R. H a y e s , D.D.S.,** D. N i c o l e l l a , M.S.C.,*** and
A. F i e l d i n g , D.M.D.****
Drexel University, and Temple University, School of Dentistry, Philadelphia, Pa.

Previous kinesiological studies of the t e m p o r o m a n d i b u l a r joint (TMJ) were based

on the motion of only one or two selected points on the mandible (such as the lower
central incisor, the m a n d i b u l a r condyle). In the present study, a technique was
developed to measure, analyze, and describe the full three-dimensional kinematic
characteristics of the TMJ during any m a n d i b u l a r activity. The technique was
based on determination of the r e l a t i v e position between the mandible and the
temporal bone from m e a s u r e m e n t of the location of points on light-weight frames
rigidly attached through splints to the m a x i l l a r y and mandibular teeth. An opto-
electric kinematic d a t a acquisition system has been used to record the location of
these points. The results of the study indicate the following major a d v a n t a g e s of
this technique over previously reported kinesiological methods: (1) the technique
provides a full description of the motion of the mandible with respect to the
temporal bone, including all the six d e g r e e s of freedom associated with this motion;
(2) the description of motion in terms of joint p a r a m e t e r s enhances i n t e r p r e t a t i o n
of the d a t a by clinicians; (3) the motion of any point of interest on the mandible can
be easily derived from the d a t a and; (4) the s y s t e m provides only negligible
interference with the n a t u r a l jaw motion of the subject. It does not require head
fixation, does not a l t e r or i n t e r f e r e with the n a t u r a l occlusion, and its light weight
causes only minimal (and negligible) loading of the mandible. (J PROSTHET DENT
1991;65:833-9.)

F o r more than a century, researchers have devel-
oped techniques to monitor and analyze the kinesiology of
the temporomandibular joint (TMJ) and applied these
techniques to study the kinematics of both the normal and
the dysfunctional TMJ. 11°
Recent advances in electronic and computer technology
facilitated the development of electronic techniques to
record the motion of the mandible. The main advantages
of these techniques over previous ones were that they could
accurately record the dynamic motion of the mandible. In
addition, because of their reduced size, these devices pro-
vided little interference with the natural motion of the
mandible.
The goal of this study was to develop a method to mea-
sure, analyze, and describe the three-dimensional kine-
matic characteristics of the human TMJ. This method was
based on studies that had used a similar technique to in-
*Associate Professor, Department of Mechanical Engineering,
Drexel University.
**Professor, Department of Orofacial Growth and Development,
Temple University, School of Dentistry.
***Graduate student, Department of Mechanical Engineering,
Drexel University.
****Professor, Department of Oral and Maxillofacial Surgery,
Temple University, School of Dentistry.
vestigate the three-dimensional kinematic characteristics
of the ankle and subtalar jointsJ 1,12 According to this
technique, the complex movements of the mandible during
a variety of functional activities can be described in terms
of six independent variables that can be easily interpreted
by clinicians. These variables include three independent
rotations and three independent translations) 3
MATERIAL AND METHODS
Measuring system and testing procedure
This study of the kinematics of the human T M J was
based on the measurement of the relative position between
the mandible and the temporal bone. To obtain this mea-
surement, a three-dimensional kinematic data acquisition
system was used. This system had the capability of
measuring the position of points located on the mandible
and the position of points located on the temporal bone.
Because these bones are not directly accessible in vivo, the
actual points measured were located on lightweight frames
that were rigidly attached to the upper and lower teeth.
The kinematic system used in this study was an opto-
electric system that measured the location in space of min-
iature light-emitting diodes (LEDs). This system had the
capability of recording the motion of up to 32 LEDs with
a spatial error of less than 0.1 mm and a sampling rate of
up to 4000 samples per second. The experimental system
is shown schematically in Fig. 1. The system consisted of
the light-emitting diodes (LEDs) corresponding to the

THE JOURNAL OF PROSTHETIC DENTISTRY 833


A- ANATOMICAL REFERENCE FRAME
T - TEMPORAL FRAME
M . MANDIBULAR FRAME
I COMPUTER
Fig. 1. Schematic diagram of experimental setup. Two
cameras (CAM1 and CAM2); controller unit; and LEDs
mounted on frames A, T, and M are part of kinematic data
acquisition system. LEDs A1, A2, A3 define location of an-
atomic frame. LEDs mounted on frame T measure position
of temporal bone. LEDs mounted on frame M measure po-
sition of mandible.
Fig. 2. Viewofmeasuringframesattachedthroughsplints
to subject's teeth.
measured points, two special cameras used to detect the
location of these points, a control unit for obtaining the in-
formation from the cameras and controlling the lighting of
the LEDs, and a microcomputer used to collect the data
and display the results.
A mandibular measuring frame M and a temporal mea-
834
S I E G L E R ET AL
Fig. 3. Schematic view of joint parameters used to de-
scribe relative position between mandible and temporal
bone. a, Angular displacement about X axis; ql, translation
along X axis;/~, angular displacement about common per-
pendicular to both X axis and z axis; q2, translation along
common perpendicular; % angular displacement about z
axis; q3, translation along z axis.
suring frame T were prepared for the subject before the
test. Each measuring frame consisted of a set of eight LEDs
embedded into a lightweight Plexiglas frame. The frames
were attached to splints prepared from dental casts of the
subject's teeth. A view of the measuring frames is shown in
Fig. 2. Once the frames were attached to the subject's teeth,
they were assumed to represent a rigid extension of the
bones articulating at the TMJ (either the mandible or the
temporal bone).
The subject was seated in front of the cameras and the
two measuring frames were attached through the splints to
the upper and lower teeth (Fig. 1). The mandibular
condyles were then identified by palpation and marked on
the skin. A head frame A was then attached to the subject's
head in such a way that two of the infrared markers (A1 and
A2) were aligned with the mandibular condyles while the
third marker (A3) was located midway between these two
points (Fig. 1). The three points defined an anatomic ref-
erence frame relative to which the motion was described.
Following these preparations, the subject was instructed
to sit motionless with his teeth in centric occlusion while
the position of the points on the mandibular frame, the
temporal frame, and the head frame were recorded through
the kinematic system. This calibration test was conducted
to establish the relationship between the measuring frames
M and T and the anatomic frame A.
After the calibration test described, the subject was
instructed to perform a series of mandibular excursions in-
cluding (1) open and close cycles; (2) protrusion and retru-
sion cycles; (3) lateral and medial cycles; (4) border move-
J U N E 1991 V O L U M E 65 NUMBER 6
T H R E E - D I M E N S I O N A L KINESIOLOGY
Fig. 4. Schematic view of six displacements of mandible where each displacement is pro-
duced by change in single joint parameter.
ments (envelope of motion) in the sagittal, coronal, and
frontal planes; and (5) chewing on both the left and right
sides on a variety of standardized food items (such as cubes
of carrots, chewing gum). During each movement, the po-
sition of the markers on the measuring frames were
continuously recorded and stored on the computer for fur-
ther analysis.
Data analysis
The analysis consisted of deriving the relative position
and orientation of the mandible with respect to the tempo-
ral bone from measurement of the coordinates of points on
the temporal and mandibular frames. The algorithm used
for this purpose was described in previous reports. 11, z2This
algorithm was based on a constrained optimization proce-
dure using a least square error criterion to find the best ap-
proximation for the relative position of the mandible with
respect to the temporal bone from measurement of the po-
sition of points on these bones. The description of the rel-
ative position was based on the joint parameters developed
THE J O U R N A L OF P R O S T H E T I C D E N T I S T R Y
by Grood and Suntay 13 after adaptation for the TMJ. The
joint parameters consist of three rotations and three trans-
lations as shown in Fig. 3. In this figure the joint parame-
ters a, 8, and ~/correspond to the three rotations and ql,
q2, and q3 correspond to the three translations.
Fig. 4 provides a schematic view of six displacements of
the TMJ where each displacement is produced by a change
in a single joint parameter. Also shown in Fig. 4 are the sign
conventions for the joint parameters.
RESULTS
To demonstrate the capabilities of the technique devel-
oped in this study, the experimental procedure described
previously was applied to study the kinesiology of the TMJ
in three young adult subjects. These subjects had no his-
tory of TMJ disorders and had normal, class I occlusions.
The mandibular movements investigated were (1) open
and close cycles; (2) protrusion and retrusion cycles; (3)
medial and lateral cycles; (4) border movements in the
sagittal, coronal, and frontal planes; and (5) chewing.
835
 SIEGLER ET AL

OPEN & CLOSE LATERAL MEDIAL MOVEMENT

(3 cycles) (I cycle)

1' 1
b~a

O"

-&
tt
n ~
',,., :-
i ),~ i

"1o
-8
-15.

-5,

-2(~

-25
.... t ' " ' ~ ' " T " T " ' ~ ' " T " T " T ' " J " ' ~ " W " ~ ' ~ ' % " ~ ' , I " ~ t ~ -10 .... ~ .... $ .... ~...
time (sec) time (sec)
10-
4° 1 . ,~, ,~

• 1.;', ,
: , , , q3
t i ! f l

) t % i t s • !
s t i i j • u
o ~ I' t J i
E 10 i
/ ql ',t ,
/ ~t s
/ ;i
# I , l e t .

-,!
q2

20 ~ , , L
-lo . . . . t . . . . t . . . . ~ . . . . ~, , . .
A time (sec) B time (sec)

Fig. 5. Joint parameters as function of time for following Fig. 5. B, Translation to right, starting from centric oc-
mandibular movements: A, open-close; B, lateral-medial; clusion, followed by translation to left and then back to
C, protrusion-retrusion. A, Three open-close cycles. centric occlusion.

Typical data obtained from one of the tested subjects are was a clear lateral deviation of the mandible to the left to
shown in Fig. 5. In Fig. 5 the joint variables (Figs. 3 and 4), a maximum of approximately 6 mm. As the mouth closed,
including the three angular displacements a,/~, ~. and the the reverse occurred. That is, a returned to 0 degree, the
three linear displacements ql, q2, q3, are shown as a func- mandible moved upward, and, at the same time, the man-
tion of time for three different activities. Fig. 5, A shows the dible deviated laterally to the right.
results obtained from three open-and-close movements Lateral deviation of the mandible (Fig. 5, B) to the right
starting from centric occlusion. Fig. 5, B corresponds to a corresponds to an axial rotation of the mandible to the right
motion of translation of the mandible to the right, then to ('y) and a rotation in which the right condyle is moving up-
the left, and back to centric occlusion. Fig. 5, C shows the ward while the left condyle is moving downward (B). Also,
joint variables corresponding to three cycles of protrusion- during this motion there is a small displacement of the
retrusion movements. mandible laterally. The motion is performed with a slight
As can be seen from Fig. 5, A, the motion of opening the protrusion of the mandible (q2 < 5 mm). Also, as the man-
mouth corresponds to an increase in a to a maximum of dible rotates to the right it translates slightly downward
approximately 20 degrees. There is little change in the (q3). The reverse occurs on motion to the left side.
other angles. At the same time, it can be seen that the Three successive cycles of protrusion-retrusion move-
mandible is moving downward (q3) to a maximum of ap- ments are shown in Fig. 5, C. As the mandible moves from
proximately 3.5 centimeters and forward (q2) to approxi- centric occlusion to maximal protrusion of q2 ffi 8 mm, it
mately 1.5 cm. Of interest, for this particular subject there also moves downward to a maximum of q3 = 6 mm. All an-

836 JUNE 1991 VOLUME 65 NUMBER 6

THREE-DIMENSIONAL KINESIOLOGY

PROTRUSION & RETRUSION (ql = -.10 mm). In this position, the angle ~ is negative
(3 cycles)
(B = - 1 degree) and the angle ~/is positive (3, = I degree).

il That is, the mandible is rotated such that the right condyle
moves upward, the left condyle moves downward, and the
mandible is axially rotated toward the chewing side. Dur-
ing the first chewing stroke, as the mouth closes, the man-
dible translates toward the position of centric occlusion (q 1
increases), the mandible rotates axially toward the neutral
position (3" decreases), and the right condyle moves down-
_, " " "~ , , ,.~ ..r" ',, :.
ward as the left condyle moves upward (~ increases). In
• , .... ,~.~ ,, , , , ,, ,.,
tt ~ ii • preparation for the second masticatory stroke, the reverse
_: v gamma V
movement occurs. That is, the mouth opens, the mandible
-3 translates toward the chewing side, and the mandible ro-
tates axially to the right.
-1
Another feature that can be observed in Fig. 7 is the pro-
time (sec)
gressive crushing of the food bolus. Because the food that
was chewed is relatively hard, the first few chewing strokes
cause the pr!mary crushing of the food item. That is, dur-
f\., ing these first few strokes, the height at which the molars
close on the piece of carrot and encounter resistance
decreases progressively. This is clearly indicated in Fig. 7
/ /
( c r l , c r 2 . . . . ) in which it is shown that during the first three
~ o,
. ..v.,/ ql

.I chewing strokes, the height at which the crushing of the

-10 .... ~ .... ~ .... ~ .... ~ ....
time (sec)
C
Fig. 5. C, Three protrusion-retrusion cycles.
gular rotations associated with this motion are small (<3
degrees). Maximal retrusion of the mandible is smaller
than maximal protrusion.
One major advantage of the experimental and analytical
technique developed in this study is that the data contain
sufficient information to calculate the motion of any point
of interest on the mandible. As demonstrated in Fig. 6, the
motion of a point of a lower central incisor is traced during
a sagittal border movement.
The results of a chewing test are shown in Fig. 7. In this
test, the subject was instructed to place a 1 cm 3 piece of
carrot between the right molars and to chew on the right
side. The joint parameters associated with this activity
were derived from the collected data and the first seven
chewing strokes are shown in Fig. 7.
The results shown in Fig. 7 indicate the following char-
acteristics associated with this chewing activity. At the be-
ginning of the chewing movement, with the piece of carrot
held between the teeth, the mouth is slightly open (q3 = 18
mm) and the mandible is translated slightly to the right
food item occurs decreases. From that cycle and until
swallowing occurs, this height remains approximately con-
stant.
A
DISCUSSION
~ .... ~ .... .~ . . . . ~ .... The most basic kinesiologic feature of the human T M J
is that this joint, unlike other joints of the skeletal system,
possesses all of the six degrees of freedom associated with
the unconstrained motion of a rigid body. Full use of these
degrees of freedom is made during diverse daily activities
such as chewing, swallowing, and speaking. This complex
movement characteristic of the T M J provides a major
challenge to scientists and clinicians investigating the
function of this joint. The challenge is to develop a
technique that, on the one hand, can measure the complex
movements occurring at this joint and, on the other hand,
can represent these complex movements in terms of easily
interpretable variables that can provide clear insight into
the characteristics of its motion. This study was directed
toward this challenge.
In the present study, a method was developed to mea-
sure, analyze, and describe the three-dimensional kine-
matic characteristics of the human TMJ. The ability of this
technique to provide a clear insight into the motion char-
acteristics of the T M J have been demonstrated through a
series of experiments conducted on three adults with nor-
mal, class I occlusions. The movements considered in this
study included open-close, protrusion-retrusion, medial-
lateral, border movements, and chewing. However, the
technique can be applied to any other mandibular activity
of interest. The present study does not provide a compre-

T H E J O U R N A L OF P R O S T H E T I C D E N T I S T R Y 837
 S I E G L E R ET AL

SIDE VIEW
--50
a~rt

-7(

~.-,o
N

-110

-130
-1 ....... :~'ob. . . . . . . :~ ....... ~0
Y (,~)

FRONT VIEW TOP VIEW

-7O

-110
t }

-130
....... -i~ ........ 6 ........ '6' ........ ....... "~b ........ 6 ........ '~b' . . . . . . .
x (.~..) x Cram)

Fig. 6. Motion of lower central incisor during sagittal border movement of mandible cor-
responds to motion of lower incisor in sagittal, frontal, and coronal planes.

hensive investigation of the kinematic characteristics of the
TMJ. The experiments and the data derived from them
were produced for the purpose of demonstrating the capa-
bilities of the technique. These capabilities can easily be
identified from the results reported in the previous section.
Some, but by no means all, of these capabilities include the
following.
1. Determination of the full range of motion of the T M J
requires determination of the maximal range of variation
in each of the six degrees of freedom of the TMJ. These can
easily be derived from the processed data corresponding to
open-close, protrusion-retrusion, medial-lateral, and bor-
der movements. An example of the data from which such
information can be derived is shown in Fig. 5.
2. The processed data derived through the use of the
technique developed in this study contain sufficient infor-
mation to determine the motion of any point of interest on
the mandible. Consequently it is possible to determine the
motion of points such as a specific tooth, the mandibular
condyles, and the interaction between teeth during the
movement. The results shown in Fig. 6 demonstrate this
capability by describing the motion of a lower central in-
cisor tooth during sagittal border movements.
3. The level of symmetry during controlled mandibular
movements can be derived from the processed data. For
example, the data shown in Fig. 5 indicate a significant lat-
eral mandibular deviation on opening and closing that may
be indicative of T M J dysfunction.
4. The technique can be used to provide a clear quanti-
tative description of the kinesiology of mastication and
other activities such as swallowing or speech. Fig. 7 dem-
onstrates this capability for a chewing activity. Timing and
magnitude information, cycle-to-cycle variation, and the
progressive crushing of food can all be easily derived from
the processed data.
SUMMARY
The main capabilities and advantages of this technique
over previous kinesiological techniques are (1) the tech-
nique provides a full description of the motion of the man-
dible with respect to the temporal bone, including all the
six degrees of freedom associated with this motion; (2) the
description of motion in terms of joint parameters en-
hances the interpretation of the data by clinicians; (3) the
motion of any point of interest on the mandible can be eas-
ily derived from the data; and (4) the system provides only

838 J U N E 1991 V O L U M E 65 NUMBER 6

THREE-DIMENSIONAL KINESIOLOGY

CHEWING MOTION (R)

I ii
0 1 2 3 4 5 6
time ( s e e )

30

CR1

CR3
E

-10

-20 i i i | i i i i i i i I i | i i i i i |1 i i ii i i i i i i i | i i i i i i i ii i

0 1 2 3 4. 5
Ume (a*¢)
Fig. 7. Joint parameters as function of time corresponding to first seven chewing strokes
associated with chewing on right side at 1 cm 3 of carrot. CR1 through CR5 represent points
during chewing stroke where teeth come in contact with item of food.

negligible interference with the natural jaw motion of the 9. Knap FJ, Richardson BL, Bogstad J. Study of mandibular motion in six
degrees of freedom. J Dent Res 1970;49:289-92.
subject. It does not require head fixation; does not alter or
10. Gibbs CH, Messerman T, Reswick JB, Derda HJ. Functional move-
interfere with the natural occlusion, and its light weight ments of the mandible. J PROSTHET DENT 1971;26:604-20.
causes only minimal and negligible loading of the mandi- 11. Siegler S, Chen J, Schneck CD. The three dimensional kinematics and
flexibility characteristics of the human ankle and subtalar joints. Part
ble.
1: kinematics. ASME Trans J Biomech Eng 1988;110:364.
This technique is presently being used by the authors to 12. Siegler S, Chen J, Bahar LY. Estimation of the general finite displace-
investigate the three-dimensional kinematic characteris- ments of a rigid body from an arbitrary number of noisy observations.
Minrmapolis, Minn: 1988 U.S.A.-Japan Symposium on Flexible Auto-
tics of the normal and the dysfunctional human temporo-
mation, 1988.
mandibular joint. 13. Grood ES, Suntay WJ. A joint coordination system for the clinical de-
scription of three dimensional motions: application to the knee. ASME
REFERENCES Trans J Biomech Eng 1983;105:136.

1. Sonstebo HR. C. E. Luce's recording of mandibular movements. J Reprint requests to:

PROSTHET DENT 1961;11:1068-73. DR. SORIN SIEGLER
2. Hilderbrand GY. Studies in the masticatory movements of the human DEPT. MECHANICALENGINEERING
lower jaw. Scand Arch Physiol 1931;(Suppl 61). DREXEL UNIVERSITY
3. Frank L. A report on the normal movement of the condyle. Dent Digest 32ND AN[) CHESTNUT STS.
1948;54:540-2. PHILADELPHIA, PA 19104
4. Posselt U. Studies in the mobility of the human mandible. Acta Odon-
tol Scand 1952;10(Suppl 10):19-160.
5. Berry HM, Hofman FA. Cineradiographic observations of temporo- Contributing authors
mandibular joint function. J PROSTHET DENT 1959;9:21-33.
6. Beck HO, Morrison WE. A method for reproduction of movements of H. Kim, M.Sc., Graduate student, Department of Me-
the mandible. J PROSTHET DENT 1962;12:873-83. chanical Engineering, Drexel University, Philadelphia, Pa.
7. Hickey JC, Allison ML, Woelfel JB, Boucher CO, Stacy RW. Mandib- S. Slawek, D.D.S., Graduate student, Department of
ular movements in three dimensions. J PROSTHET DENT 1963;13:72-92.
8. Messerman T. A means for studying mandibular movements. J PROS- Orofacial Growth and Development, Temple University,
THET DENT 1967;17:36-43. School of Dentistry, Philadelphia, Pa.

T H E J O U R N A L OF P R O S T H E T I C DENTISTRY 839