Professor Uchida and the CZI funding for open source software: moving toward success

Sculptures showing movements

A step in the right direction: Biomechanics researcher from uOttawa receives funding from the Chan Zuckerberg Initiative to develop open-source software tools that can improve our understanding of how we move. Professor Uchida, an Assistant Professor in Mechanical Engineering, is one of the grant recipients in Cycle 3 of the Essential Open Source Software for Science program. But what does this grant mean from his perspective and how will it impact the projects he is working on? This article will answer these questions and describe the positive effect these funds will have on his research.

Why do we study movement?

Movement, emerging through the interaction of complex neural, muscular, and skeletal systems, is fundamental to human and animal life. Experiments go a long way in helping us understand how we move, but software tools allow us to compute quantities that are difficult or impossible to measure directly. By generating simulations of musculoskeletal models that are driven by experimental data, we can estimate the force that each muscle must generate to produce movements like walking and running. We can then compute the amount of energy stored and released by our tendons as they stretch and recoil, the load in the knee of elderly individuals with osteoarthritis, and many other variables. In short, looking “under the hood” gives us insight into how our biological systems work and provides directions for designing more effective treatments and assistive devices.

What is Professor Uchida currently working on?

Since 2012, he has been part of the Development Team for OpenSim, an open-source software platform for modelling musculoskeletal systems and generating simulations of human and animal movement. The software has thousands of users worldwide and it has been used in approximately 1,500 published studies of movement. Most of these studies have been performed by researchers who spent months learning to use the software effectively. To make OpenSim’s capabilities available to a broader community of clinicians planning treatments, engineers designing assistive devices, roboticists building exoskeletons, and others, the modelling and simulation process must be simpler.

About OpenSim: what is the focus of this project?

This project will automate model calibration, the procedure by which a musculoskeletal model is adjusted to match the dimensions of a human subject—a notoriously difficult process for new users. Model calibration is a critical first step in any analysis as the quality of the subsequent simulation relies heavily on providing a properly calibrated model. In the current process, the OpenSim graphical user interface is used to position spherical markers on a generic model to match the anatomical locations of optical markers that were affixed to a subject’s skin during an experiment. The dimensions of the model’s body segments are then adjusted to reduce the distance between each model marker and the corresponding experimental marker from one frame of video. This process is tedious and iterative because there is uncertainty in where the markers should be placed on the model. Furthermore, the optical markers move relative to the underlying bones we wish to track as the skin stretches during movement, so it is often necessary to fine-tune the marker locations to reduce error.

How will the funding impact this project?

The provided funding of $130,000 USD will support one software developer and one OpenSim usability specialist. They will work with Prof. Uchida at uOttawa and the rest of the OpenSim Development Team at Stanford to design, develop, test, and disseminate a new optimization method that improves the model calibration process, obtaining better models in less time and without relying on the experience of the user. The project is part of the Chan Zuckerberg Initiative’s Essential Open Source Software for Science program, which supports open-source software projects “to help make the computational foundations of biological research more usable and robust.”


For more information or inquiries about the open positions supported by this project, please contact Thomas Uchida. More information about his work can be found on his website.

 

Back to top