Remember to update your bookmarks with our new URL.
If you find missing content or broken links, let us know: opensim@stanford.edu

Evaluating a Musculoskeletal Model

Getting Started: Defining your Research Goals for the Model

  • What results do I need to answer my research question?
  • What elements of a model might these results be sensitive to?
  • In what ranges will my study be conducted and in what ranges has the model been validated? If a model is not valid in the ranges of your study, how can you extend those ranges and re-validate the model?

General Questions

  • What structures or systems are modeled?
  • Is it a kinematic model or full dynamic model? Is it 3D?
  • Is it a model of a new structure or structures? Does it build on an existing model? What models can it be compared to?
  • What experimental data was used to build and test the model? If human or animal subjects were used, how many and what type of subjects (age, size, etc.)?
  • What platform(s) is the model implemented in?
  • Can the model be scaled or otherwise customized?
  • Has the model been published or otherwise peer reviewed?
  • Do other studies use/cite the model?
  • Has the author of the model changed or updated the model since the original publication?  Do I have and understand the most recent version?

Kinematic Models

Musculoskeletal Geometry

  • Bone Geometry
    • What bony segments are included in the model?
    • What experimental or literature data was used to define bone geometry?
  • Joint Geometry
    • How many joints and how many degrees of freedom?
    • How are the joints modeled? Types? Joint axis definitions? Coordinates? Neutral positions?
    • Is there coupling of joint motions?
    • What is the joint range of motion? Are there other limits or other constraints?
    • What experimental or literature data was used to define the joint geometry?
  • Muscle Geometry
    • How many muscles are included in the model?
    • How are muscle attachment points defined?
    • Are broad muscles represented by multiple segments?
    • Does the model have wrapping surfaces or via points? What features do they represent?
    • What experimental or literature data was used to define the muscle geometry?
  • Other Structures
    • Does the model include ligaments?
    • Does the model include prosthetics, orthotics, or other devices?

Dynamic Models

Inertial Properties

  • Does the model include segment inertial parameters?
  • What experimental or literature data was used to define the intertial parameters?
  • Are there segments without intertial parameters?  (This is a challenge for creating forward simulations with the model.)

Actuators and Other Force-Generating Elements

  • Does the model include muscle actuators?
    • What type of model is used for muscle-tendon force generation?  
    • How is muscle architecture modeled? What parameters define muscle architecture? 
    • How is activation-contraction dynamics of muscles modeled? What parameters define?
    • Was there any fine-tuning/adjustment of muscle model parameters?
  • Does the model have any other actuators (e.g. powered prosthetics, reserve actuators at the joints)?
  • Does the model have any other passive force-generating elements (e.g. damping at joint extremes)?

Controllers and Sensors

  • Does the model have any custom controllers?
  • Does the model have any custom sensors?


  • Is a model of foot/floor contact included?
  • Does the model include any other representations of of contact with external objects?

Model Testing

  • Were model moment arms compared to experimental data?
  • Were maximum active joint moments compared to experimental data?
  • Were passive joint moments compared to experimental data?
  • Was a sensitivity analysis performed for any of the parameters of the model?
  • Has the model been used to generate simulations?
    • How much time is required to generate simulations of standard motions (e.g. walking, reaching)?
    • How do the results compare to previously published data?
    • For example, if a lower extremity model, does the model accurately predict knee forces from the grand challenge data set (https://simtk.org/home/kneeloads)?

More Help

  • The webinar “Musculoskeletal Models Deconstructed” provides some specific examples of how to evaluate and develop a model (watch | view slides).

OpenSim is supported by the Mobilize Center , an NIH Biomedical Technology Resource Center (grant P41 EB027060); the Restore Center , an NIH-funded Medical Rehabilitation Research Resource Network Center (grant P2C HD101913); and the Wu Tsai Human Performance Alliance through the Joe and Clara Tsai Foundation. See the People page for a list of the many people who have contributed to the OpenSim project over the years. ©2010-2024 OpenSim. All rights reserved.