Journal of Biomechanics 31 (1998) 963 — 967

Technical Note
An analytical model of the knee for estimation of internal forces
during exercise
Naiquan Zheng, Glenn S. Fleisig*, Rafael F. Escamilla, Steven W. Barrentine
American Sports Medicine Institute, 1313 13th Street South, Birmingham, AL 35205, U.S.A.
Received in final form 13 March 1998

Abstract

An analytical model of the knee joint was developed to estimate the forces at the knee during exercise. Muscle forces were estimated
based upon electromyographic activities during exercise and during maximum voluntary isometric contraction (MVIC), physiological
cross-sectional area (PCSA), muscle fiber length at contraction and the maximum force produced by an unit PCSA under MVIC.
Tibiofemoral compressive force and cruciate ligaments’ tension were determined by using resultant force and torque at the knee,
muscle forces, and orientations and moment arms of the muscles and ligaments. An optimization program was used to minimize the
errors caused by the estimation of the muscle forces. The model was used in a ten-subject study of open kinetic chain exercise (seated
knee extension) and closed kinetic chain exercises (leg press and squat). Results calculated with this model were compared to those
from a previous study which did not consider muscle length and optimization. Peak tibiofemoral compressive forces were
3134$1040 N during squat, 3155$755 N during leg press and 3285$1927 N during knee extension. Peak posterior cruciate
ligament tensions were 1868$878 N during squat, 1866$383 N during leg press and 959$300 N for seated knee extension. No
significant anterior cruciate ligament (ACL) tension was found during leg press and squat. Peak ACL tension was 142$257 N during
seated knee extension. It is demonstrated that the current model provided better estimation of knee forces during exercises, by
preventing significant overestimates of tibiofemoral compressive forces and cruciate ligament tensions. ( 1998 Elsevier Science Ltd.
All rights reserved.

Keywords: Knee model; Muscle force; Cruciate ligament; Knee exercise

1. Introduction Recently, resultant forces of the knee during differ-

ent exercises have been studied (Lutz et al., 1993; Stuart
Understanding the biomechanics of the knee during et al., 1996; Wilk et al., 1996). Some efforts were made
exercises, such as knee extension, leg press and squat, is in these studies to estimate muscle forces and to com-
very important for therapists to design rehabilitation pare muscle activity; however, further study of tensions
programs and for trainers to strengthen and condition of cruciate ligaments and the bone-to-bone contact
athletes. Muscle force, ligament tension and joint contact force of the tibiofemoral joint was prevented by limits
forces between the femur and tibia during exercise are the of the biomechanical models utilized. Muscle forces
main issues to evaluate exercises. For example, therapists were estimated based only on the physiological cross-
try not to overload the anterior cruciate ligament (ACL) sectional area and electromyographic activities, not
for ACL injured and reconstructed patients during their muscle fiber length at contraction. Thus, muscle forces
rehabilitation programs. It has been controversial which were overestimated at full extension for the knee exten-
exercise should be utilized for a patient since the ACL sors without taking into account the shorter muscle fiber
tension during different exercises are not fully under- length.
stood (Wilk et al., 1997). The purpose of this study was to develop an analytical
model of the knee in the sagittal plane during quasistatic
(i.e. isometric or low speed) exercise. The tibiofemoral
compressive force and tensions of the anterior and poste-
* Corresponding author. Tel.: 001 205 918-2138; fax: 001 205 918- rior cruciate ligament were presented during the knee
0800; e-mail: [email protected] extension, leg press and squat.

0021-9290/98/$19.00 ( 1998 Elsevier Science Ltd. All rights reserved.

PII: S 0 0 2 1 - 9 2 9 0 ( 9 8 ) 0 0 0 5 6 - 6
964 N. Zheng et al. / Journal of Biomechanics 31 (1998) 963—967

2. Materials and methods cross-bridge model of muscle. The muscle length was
determined by the knee joint angle, hip joint angle and
Resultant force and moment at the knee were the sums muscle line of action and geometry of the lower ex-
of those due to individual ligament, muscle and joint tremity (Pierrynowski, 1991), which determined the
contact: muscle fiber length and sarcomere length (Herzog et al.,
1990). For single joint knee muscles, the muscle
n. nl n# length factor k was determined as a segmental linear
F " + F # + Flj# + F , (1)
3%4 .i #k function of knee flexion angle since the isometric tension
i/1 j/1 k/1
of the muscle is a segmental linear function of the sar-
n. nl n# comere length (Gordon et al., 1966). For two joint
M " + M # + Mlj# + M , (2)
3%4 .i #k muscles, k was a segmental linear function of both knee
i/1 j/1 k/1
and hip flexion angles (Fig. 1). Since the model was
M "r ]F (A"m, l, c; i"1, 2, 3, 2 , n ), (3)
Ai Ai Ai A developed for quasistatic exercise, no force-velocity fac-
where F and M were the resultant force and moment tor was included.
3%4 3%4 PCSA data from Wickiewicz et al. (1983) were used to
at the knee; n , nl and n were the number of muscle
. # determine the ratios of PCSA between muscle groups.
groups, ligament bundles and bone contact interfaces
used in the model; F and M were the force and According to Narici et al. (1988), the total PCSA of the
.i .i quadriceps (PCSA ) was approximately 160 cm2 for
moment of the ith muscle group; and Fl and Ml were the
j j
Q
force and moment of the jth ligament bundle; F and a 75 kg man. Total PCSA of the quadriceps was scaled
#k up or down by individual body mass (PCSA "
M were the force and moment of the kth bone contact
#k Q
element; and r was the vector of a moment arm. 160 * BW/75 cm2). The maximum voluntary contraction
Ai force per unit PCSA was assumed to be 40 N cm~2 for
To simplify the model some assumptions were made:
(a) collateral ligaments were not included since they have the quadriceps and 35 N cm~2 for the hamstring and
little effects on the mechanics of the knee in the sagittal gastrocnemius muscles (Narici et al., 1988; Cholewicki et
plane (Crowninshield et al., 1976), (b) the flexion—exten- al., 1995; Narici et al., 1992; Wickiewicz et al., 1984).
sion moment of the cruciate ligaments were ignored since Moment arms of muscle forces and angles of the line of
they were located close to the rotation center of the knee, action for muscle and ligaments were represented as
(c) the tibiofemoral force was assumed to be applied in polynomial functions of the knee flexion angle using data
the line of knee rotation center, therefore, the contribu- from Herzog and Read (1993) (Tables 1 and 2). Since lines
tion of the tibiofemoral force to the resultant extension of action and moment arms for muscles and ligaments
torque was ignored; and (d) the tibiofemoral joint and from literature were limited to the sagittal plane, only the
patellofemoral joint were assumed to be frictionless due forces in the sagittal plane were analyzed in this model.
to the synovial fluid. Using the equilibrium equations the cruciate ligament
forces (F , F ) and tibiofemoral force (F ) were
In this model, resultant forces and torques at the knee PCL ACL TF
were assumed to be known. Forces of the muscles were
estimated using surface EMG data from eight muscle
groups (rectus femoris, vastus medialis, vastus lateralis,
vastus intermedius, biceps femoris, semitendinosus,
semimembranosus, and gastrocnemius) and the following
equation:

EMG
F "c k A p i, (4)
.i i i i .i MVC
i
where k was a muscle force—length factor defined as
i
a function of knee and hip flexion angle; A was the
i
physiological cross-sectional area (PCSA) of the ith
muscle; p was maximum voluntary isometric contrac-
.i
tion force per unit PCSA of the muscle; EMG and
i
MVIC were window averages of ith muscle EMG during
i
exercise and maximum voluntary isometric contraction;
coefficient c was the weigh factor which was adjusted in
i
an optimization program to minimize the errors in
muscle force estimation. Fig. 1. Force—length factor k for two joint muscle of quadriceps vs knee
The muscle force—length factor k was determined by and hip flexion angle. Muscle fiber length has significant effect on the
the knee joint angle and hip joint angle based on the muscle force output.
 N. Zheng et al. / Journal of Biomechanics 31 (1998) 963—967 965

Table 1
Muscle moment arm lengths (¸, m) as a polynomial function of knee angle (h, degree) based on Herzog and Read (1993)

Function ¸"B #B h#B h2#B h3#B h4

0 1 2 3 4
Coefficient B B B B B
0 1 2 3 4
Patellar tendon 4.71E-02 4.20E-04 !8.96E-06 4.47E-08 0.00
Biceps femoris 1.46E-02 !9.26E-05 8.55E-06 !8.78E-08 2.38E-10
Semimembranosus 2.84E-02 !1.61E-04 6.81E-06 !8.80E-08 2.77E-10
Semitendinosus !4.11E-03 5.86E-04 6.90E-06 !5.31E-8 0.00
Gastrocnenius 1.99E-02 !3.50E-04 9.20E-06 !1.03E-07 4.07E-10

Table 2
Lines of action for muscles and ligaments (/, degree, 0 to the anterior, 90 to the inferior) as a polynomial function of knee angle (h, degree, 0 for full
extension) based on Herzog and Read (1993)

Function /"B #B h#B h2#B h3

0 1 2 3
Coefficient B B B B
0 1 2 3
Patellar tendon !74.40 !5.75E-02 !4.75E-03 3.09E-05
Biceps femoris 275.00 !0.8720 !7.12E-04 0.00
Semimembranosus 260.00 !0.8880 !8.52E-04 0.00
Semitendinosus 255.00 !0.8160 2.63E-04 !6.19E-6
Gastrocnenius !90.00 0.00 0.00 0.00
Anterior cruciate ligament 227.00 !0.4880 0.00
Posterior cruciate ligament !66.00 0.7370 !4.96E-03 0.00

estimated. Eq. (1) can be rewritten as where c and c were the lower and upper limit for
-08 )*')
c and j was a constant. The coefficient c was set at one at
n. i i
F #F #F "F ! + F . (5) beginning and adjusted by using the Davidon—Fletcher—
TF PCL ACL 3%4 .i Powell algorithm (Dennis and Schnabel, 1983).
i/1
To test the model, previously collected data were used
When the resultant anteroposterior shear force was pos-
(Wilk et al., 1996). A three-dimensional motion analysis
itive, the tension of the posterior cruciate ligament was
system (Motion Analysis Corporation, Santa Rosa, CA,
determined and tension of the anterior cruciate ligament
U.S.A.) and force platform (Advanced Mechanical Tech-
was assumed to be zero. Conversely, when the resultant
nology, Inc., Newton, MA, U.S.A.) were used to collect
anteroposterior shear force was negative, tension of the
kinematic data, external loads and ground reaction for-
anterior cruciate ligament was determined and tension of
ces for ten subjects performing three exercises: seated
the posterior cruciate ligament was assumed to be zero.
knee extension, leg press and squat. The knee flexion or
The patellar tendon force was determined by the quad-
extension angular velocities were found to be approxim-
riceps tendon force and the ratio of the patellar tendon
ately 90° s~1 for peak value and approximately 60° s~1
force to the quadriceps tendon force (Van Eijden et al.,
for average, making data acceptable for the quasistatic
1987).
assumption in the model. The resultant force and mo-
Since the accuracy of the estimation of the muscle
ment at the knee were determined by using three-
forces depends on the accuracy of the estimation of the
dimensional rigid link models and principles of inverse
physiological cross-sectional area, maximum voluntary
dynamics (Feltner and Dapena, 1989; Wilk et al., 1996).
isometric contraction force per unit PCSA, EMG data
Surface EMG data were collected from rectus femoris,
and muscle fiber length, coefficient c was used for each
i vastus lateralis, vastus medialis, biceps femoris, medial
muscle in estimation of its force. These coefficients were
hamstrings and gastrocnemius muscles with a Noraxon
determined in an optimization program by minimizing
system (Noraxon, Pheonix, AZ, U.S.A.). The average of
the difference between the resultant moment of the knee
the rectus femoris, vastus lateralis and vastus medialis
from kinetic analysis (M ) and that from the estimation
3%4 EMG data was used to represent the EMG data of the
of the model (+ M ). The following objective function
.i vastus intermedius. EMG data of the medial hamstrings
was used:
were used for both the semitendinosus and semimem-

A B
n. n. 2 branosus. The window average of EMG during exercises
min f (c )" + (1!c )2#j M ! + M was expressed as a percentage of the window average of
i i 3%4 .i
i/1 i/1 EMG during the maximum voluntary isometric contrac-
subject to c )c )c , (6) tion, which was recorded when the muscle force—length
-08 i )*')
966 N. Zheng et al. / Journal of Biomechanics 31 (1998) 963—967

factor was one, and then the average for three repetitions
of each exercise was taken based on the knee angle. In
order to consider the electromechanical delay window
averages were calculated only for past history with a win-
dow width of 100 ms. The c , c and were set to 0.5,
-08 )*')
1.5 and 1.0, respectively, for optimization.

3. Results

The tibiofemoral compressive force during leg press

and squat increased with knee flexion angle, while the
tibiofemoral compressive force during knee extension
increased with decrease of knee flexion angle (Fig. 2).
Peak tibiofemoral forces were 3134$1040 N during
the squat, 3155$755 N during the leg press and
3285$1927 N during the knee extension. Fig. 2. Tibiofemoral compressive force and PCL/ACL tension vs knee
There were no significant tension in the anterior cruci- flexion angle during knee extension (circle), leg press (triangle) and
ate ligament during the leg press and squat (Fig. 2). The squat (square). There were no significant difference in peak tibiofemoral
compressive forces among three exercises, but there were significant
tension in the anterior cruciate ligament increased when difference in peak ACL and PCL tensions.
the knee approached full extension during the seated
knee extension with a peak value of 142$257 N. In
general, tension in the posterior cruciate ligament in-
creased with knee flexion. Peak PCL tensions were
1868$878 N during the squat, 1866$383 N during the
leg press and 959$300 N during the seated knee exten-
sion. Optimization produced significant adjustments on
muscle forces (Fig. 3).

4. Discussion

A planar quasistatic model of the knee in the sagittal

plane was developed, which can be expanded to three-
dimensional when lines of action and moment arms of
ligaments and muscles are determined in three dimen- Fig. 3. Muscle force estimation of quadriceps before and after optim-
sions. The model was used for a study of 10 subjects ization for a squat of one subject. Optimization had significant adjust-
performing the knee extension, leg press and squat ments on muscle forces.
(Zheng et al., 1996). The model took into account muscle
activities, especially the co-contraction of the extensors
and flexors of the knee. Although co-contraction of the studies which did not consider muscle length and optim-
hamstrings and quadriceps may have no effect on the ization. Linear muscle activity—force relationship was
extension torque of the knee joint, it has a significant assumed at a given knee flexion angle (Bean et al., 1988;
effect on tibiofemoral compressive force and anteropos- Lawrence et al., 1983). The muscle force—length factor
terior shear force. Co-contraction of the hamstrings and k was about 0.25 for vastus lateralis, medialis and inter-
quadriceps is an important factor in determining tensions medius when the knee was in full extension. By ignoring
in the cruciate ligaments. The co-contraction has more the force—length factor the muscle forces would be esti-
effect on the tension in the posterior cruciate ligament mated as high as four times. When only EMG and PCSA
when the knee flexed 90°, and has more effect on the were considered, the tibiofemoral compressive forces
tension in the anterior cruciate ligament when the knee is were overestimated with peak values of 6139$1709 N
near full extension. This is due to varying lines of action during the squat, 5762$1508 N during the leg press and
for the patellar tendon and hamstrings muscle. Thus, it is 4598$2547 N during the knee extension (Wilk et al,,
important to take into account the co-contraction of the 1996), much higher than those calculated with this model.
hamstrings and quadriceps. By ignoring the force—length factor of the muscles, higher
The model provided better estimation of internal for- tibiofemoral compressive force (3453$1313) at 30° knee
ces of the knee during exercises than those in previous flexion was reported than that (2198$805) at 90° during
 N. Zheng et al. / Journal of Biomechanics 31 (1998) 963—967 967

isometric leg press (Lutz et al., 1993). No cruciate Dennis, J.E., Schnabel, R.B., 1983. Numerical Methods for Uncon-
ligament tensions during exercises have been previously strained Optimization and Nonlinear Equation. Prentice-Hall, En-
reported. Tibiofemoral shear forces were reported to glewod Cliff, NJ.
Feltner, M.E., Dapena, J., 1989. Three-dimensional interactions in
compare different exercises (Wilk et al., 1996; Lutz et al., a two-segment kinetic chain: Part I: general model. International
1993). Wilk et al. reported maximal posterior shear force Journal of Sport and Biomechanics 5, 403—19.
of 1783$634 N during the squat, 1667$556 N during Gordon, A.M., Huxley, A.F., Julian, F.J., 1966. The variation in isomet-
the leg press and 1178$594 N during the knee extension. ric tension with sarcomere length in vertebrate muscle fibers. Journal
After considering the orientation of the PCL at the knee of Physiology (London) 185, 170—192.
Herzog, W. Abrahams, S.K., Keurs, H., 1990. Theoretical determina-
flexion angle when these maximal shear force occurred, tion of force—length relations of intact human skeletal muscles using
PCL tensions would be about 2318 N during the squat, the cross-bridge model. European Journal of Applied Physics 416,
2167 N during the leg press and 1531 N during the knee 113—119.
extension, which were respectively 24, 16 and 60% higher Herzog, W., Read, L.J., 1993. Lines of action and moment arms of the
than those calculated with this model. major force carrying structures crossing the human knee joint. Jour-
nal of Anatomy 182, 213—230.
The optimization program was used for adjusting the Lawrence, J.H., DeLuca, C.J., 1983. Myoelectric signal versus force
muscle force calculations, based upon knee resultant relationship in different human muscles. Journal of Applied Physi-
forces and torques. EMG assisted optimization has been ology 54, 1653—1659.
recently introduced in muscle force estimation Lutz, G.E., Palmitier, R.A., An, K.E., Chao, E.Y.S., 1993. Comparison
(Cholewicki and McGill, 1994). The objective function of tibiofemoral joint forces during open-kinetic-chain and closed-
kinetic-chain exercises. Journal of Bone and Joint Surgery 75-A,
allowed the existence of co-contraction of quadriceps and 732—739.
hamstring muscles. Various objective functions from the Narici, M.V., Roi, G.S., Landoni, L., 1988. Force of knee extensor and
literature (Pedersen et al., 1987) were tried and found to flexor muscles and cross sectional area determined by nuclear mag-
be unable to include the co-contraction of the extensors netic resonance imaging. European Journal of Applied Physics 57,
and flexor of the knee. Smaller range of lower and upper 39—44.
Narici, M.V., Landoni, L., Minetti, A.E., 1992. Assessment of human
limit for c will allow less adjustment for muscle force
i knee extensor muscles stress from in vivo physiological cross-sec-
estimation and may lose some effect of optimization. tional area and strength measurement. European Journal of Applied
Smaller (0.1, 0.01) should be used if the accuracy of the Physics 65, 438—444.
known resultant extension torque of the knee (M ) is Pedersen, D.R., Brad, R.A., Cheng, C., Arora, J.S., 1987. Direct
3%4 comparison of muscle force predictions using linear and non-
lower. Further study is needed to improve the muscle
linear programing. Journal of Biomedical Engineering 109,
force—length relationship during knee exercises, espe- 192—199.
cially to specify the difference between concentric and Pierrynowski, M.R., 1991. Analytic representation of muscle line of
eccentric contractions. The model does not include action and geometry. In: Allard, P, Stokes, I., Blanchi, J. (Eds.),
a force-velocity relationship for muscle force estimation Three-dimensional Analysis of Human Movement , Human Kinetics.
since it was developed for low speed exercise. Force- pp. 215—256.
Stuart, M.J., Meglan, D.A., Lutz, G.E., Growney, E.S., An, K.N., 1996.
velocity relationships should be included in the model Comparison of intersegmental tibiofemoral joint forces and muscle
before it can be used for high-speed knee activities. The activity during various closed kinetic chain exercise. American Jour-
improvement of the model could be achieved with in- nal of Sports Medicine 24, 792—799.
creased knowledge of the assumptions used in the model Van Eijden, T.M., Kouwenhoven, E., Weijs, W.A., 1987. Mechanics of
including PCSA, EMG—force relationship, lines of action the patellar articulation. Acta Orthopaedica Scandinavica 58,
160—566.
of muscles and other data needed to expand to a three- Wickiewicz, T.L., Roy, R.R., Powell, P.L., 1983. Muscle architecture of
dimensional model. the human lower limb. Clinical Orthopedics and Related Research
179, 275—83.
Wickiewicz, T.L., Roy, R.R., Powell, P.L., Perrin, J.J., Edgerton, V.R.,
1984. Muscle architecture and force-velocity relationships in humans.
References Journal of Applied Physiology: Respiratory, Environmental and
Exercise Physiology 57, 435—443.
Bean, J.C., Chaffin, D.B., Schultz, A.B., 1988. Biomechanical model Wilk, K.E., Escamilla, R.F., Fleisig, G.S., Barrentine, S.W., Andrews,
calculation of muscle contraction forces: a double linear program- J.R., Boyd, M.L., 1996. A comparison of tibiofemoral joint force
ming method. Journal of Biomechanics 21, 59—66. and electromyographic activity during open and closed kinetic
Cholewicki, J., McGill, S.M., 1994. EMG assisted optimization: a hy- chain exercises. American Journal of Sports Medicine 24,
brid approach for estimating muscle forces in an indeterminate 518—527.
biomechanical model. Journal of Biomechanics 27, 1287—1289. Wilk, K.E., Zheng, N., Fleisig, G.S., Andrews, J.R., Clancy, W.G., 1997.
Cholewicki, J, McGill, S.M., Norman, R.W., 1995. Comparison of Kinetic chain exercise: implications for the anterior cruciate ligament
muscle forces and joint load from an optimization and EMG assisted patient. Journal of Sports Rehabilitation 6, 125—143.
lumbar spine model: towards development of a hybrid approach. Zheng, N., Fleisig, G.S., Escamilla, R.F., Barrentine, S.W., Wilk, K.E.,
Journal of Biomechanics 28, 321—31. Andrews, J.R., 1996. Forces at the knee during open and closed
Crowninshield, R, Pope, M.H., Johnson, R.J., 1976. An analytical kinetic chain exercises. Proceedings of 20th American Society of
model of the knee. Journal of Biomechanics 9, 397—40. Biomechanics Annual Meeting, Atlanta, GA, pp. 75—76.