Team Members

  • Adam Gotlin
  • Kirsten Seagers

Project Video


Medial compartment osteoarthritis is a leading cause of years claimed by disability worldwide. Joint replacements improve the quality of life for individuals with end-stage osteoarthritis; however, less invasive interventions to prevent or delay surgery are desirable. Increased contact forces in the medial compartment of the knee joint are thought to accelerate the progression of osteoarthritis.

Knee adduction moment (KAM) and total tibiofemoral force (TF) are potential correlates to medial contact force (MCF) in the knee joint (Kutzner et al., 2013; Nagura et al., 2006). Shull et al. (2012) showed experimentally that shifting the foot progression angle inward during walking (i.e. toe-in gait) could reduce the first peak in KAM during stance. This is accomplished by shifting foot's center of pressure laterally and altering the ground reaction force vector. Lowering the KAM tends to shift knee contact forces laterally, thus potentially alleviating load on the medial compartment. DeMers et al. (2014) showed that muscle recruitment optimization could be leveraged to lower the second peak of TF. His study reveals the sensitivity of contact forces in the knee to internal muscle forces, particularly for muscles crossing the knee joint. Winby et al. (2009) further describes that medial contact force is 42% external forces factors (i..e ground reaction forces) and 58% internal factors (i.e. muscle activation). Decreasing MCF is the focus of interventions to decelerate progression of medial compartment osteoarthritis.

By building a lower-body musculoskeletal model that combines Rajagopal’s full-body musculoskeletal model and Lerner’s contact model of the knee joint, we will confirm whether toe-in gait reduces medial contact force in the knee (Rajagopal et al., 2012; Lerner et al., 2015).

Research Question

Does decreasing the foot progression angle (toe-in gait) reduce medial contact loads in the knee?


Data Collection

  1. Track gait kinematics from one subject using Motion Analysis MOCAP system on an instrumented Bertec treadmill
  2. Compare gait cycles from baseline walking (N = 3 steps) to toe-in gait (N = 3 steps)
    1. Average foot progression angle for walking trials:
      1. Baseline: 6.5°
      2. Toe-in: -5.3°

Analysis in OpenSim

  1. Scale a generic musculoskeletal model to get a subject specific model

    1. Start with a combined model from Apoorva Rajagopal’s full body model and Zach Lerner’s knee contact model

    2. Modify mass in scale tool

    3. Modify tibial joint angles in the model .osim file using patient X-ray image

    4. Adjust marker positions from static trial (one leg is in front of the other)

  2. Run Inverse Kinematics

    1. Input: tracker file (<trialname>.trc)

    2. Outputs: coordinates file (<trialname_IK>.sto)

  3. Run Inverse Dynamics

    1. Inputs: coordinates file; external forces (<GRFs>.mot)

    2. Outputs: results file (<trialname>_results_ID.sto)

  4. Run Static Optimization

    1. Inputs: coordinates file; external forces

    2. Outputs: muscle forces (<trialname>_StaticOptimization_force.sto)

  5. Use Analyze Tool to calculate joint reaction forces

    1. Input: muscle forces

    2. Outputs: results file (<trailname>_ReactionLoads.sto)

      1. This will contain joint reaction forces in the compartments of the knee

  6. Repeat steps 2-4 for all walking trials



Knee Adduction Moment

Conclusion - Toe-in gait reduces first peak in knee adduction moment

Knee Flexion Moment

Conclusion - Toe-in gait reduces range of knee flexion moment

Muscle Force

Conclusion - Toe-in gait reduces quadriceps force during early stance, reduces gastrocnemius force during late stance, and has little impact on hamstring force

Joint Angles

Conclusion - Toe-in gait leads to shorter steps characterized by lower hip extension during late stance

Tibiofemoral Force

Conclusion - Toe-in gait yields decreased total knee contact force in late stance

Medial Contact Force

Conclusion: Toe-in gait reduces the first peak of medial contact force in the knee


  • Study analyzes 6 total steps for 1 subject

  • Muscle forces are estimated using static optimization

    • Joint reaction analysis is very sensitive to muscle forces

  • Model limitations

    • Lack of arm swing, trunk motion, ligaments, patch contact, etc.

  • Experiments were performed in a laboratory setting on a treadmill

Project Files

Find the model, setup files, and our joint reaction results on our simtk webpage:

Future Work

  • Perform and an EMG-driven analysis to improve accuracy of muscle forces (vs. over static optimization alone)

  • Perform a longitudinal study with a wider population to understand if kinematic and muscle recruitment adaptations persist over time

  • Analyze effectiveness of combining toe-in gait with optimized muscle control to reduce both the first and second peak medial contact force in the knee


We would like the thank the following individuals for assisting with our project

  • Scott Ulrich for posing the research question and providing guidance throughout our project
  • Julie Thompson-Kolesar, Melissa Boswell, Tom Uchida, and Scott Delp for providing direct consultation and advice
  • Apoorva Rajagopal and Zachary Lerner for constructing the baseline models we leveraged in our work


DeMers, Matthew S., et al., 2015. Changes in Tibiofemoral Forces due to Variations in Muscle Activity during Walking. J Orthop Res. 32(6): 769–776.

Lerner, Zachary F. et al., 2016. How Tibiofemoral Alignment and Contact Locations Affect Predictions of Medial and Lateral Tibiofemoral Contact Forces. Journal of Biomechanics, 48(4), 644–650.

Meghan K. Sylvia, et al., 2015. Development of a human knee joint finite element model to investigate cartilage stress during walking in obese and normal weight adults. Summer Biomechanics, Bioengineering and Biotransport Conference, 2015,Snowbird Resort, Utah, USA. 

Rajagopal, Apoorva, et al., 2016. Full-Body Musculoskeletal Model for Muscle-Driven Simulation of Human Gait. IEEE Transactions on Biomedical Engineering, VOL. 63, NO. 10

Shull, Pete B., et al., 2012. Toe-in gait reduces the first peak knee adduction moment in patients with medial compartment knee osteoarthritis. Journal of Biomechanics, 46 (2013), 122–128.

Uhlrich, Scott D., et al.. Subject-specific toe-in or toe-out gait modifications reduce the larger knee adduction moment peak more than a non-personalized approach. Journal of Biomechanics, Article In Press.

Winby CR, Lloyd DG, Besier TF, Kirk TB., 2009. Muscle and external load contribution to knee joint contact loads during normal gait. Journal of Biomechanics, 42(14), 2294-300