Biology subjects

Dembia, C. L.

Publications and source records attributed to Dembia, C. L..

3 recordsLinked to original sources

bioRxiv2019

OpenSim Moco: Musculoskeletal optimal control

Musculoskeletal simulations of movement can provide insights needed to help humans regain mobility after injuries and design robots that interact with humans. Here, we introduce Open-Sim Moco, a software toolkit for optimizing the motion and control of musculoskeletal models built in the OpenSim modeling and simulation package. OpenSim Moco uses the direct collocation method, which is often faster and can handle more diverse problems than other methods for musculoskeletal simulation but requires extensive technical expertise to implement. Moco frees researchers from implementing direct collocation themselves, allowing them to focus on their scientific questions. The software can handle the wide range of problems that interest biomechanists, including motion tracking, motion prediction, parameter optimization, model fitting, electromyography-driven simulation, and device design. Moco is the first musculoskeletal direct collocation tool to handle kinematic constraints, which are common in musculoskeletal models. To show Mocos abilities, we first solve for muscle activity that produces an observed walking motion while minimizing muscle excitations and knee joint loading. Then, we predict a squat-to-stand motion and optimize the stiffness of a passive assistive knee device. We designed Moco to be easy to use, customizable, and extensible, thereby accelerating the use of simulations to understand human and animal movement.

bioengineering↗

bioRxiv2019

Foot strike pattern during running alters muscle-tendon dynamics of the gastrocnemius and the soleus

Running is thought to be an efficient gait due, in part, to the behavior of the plantar flexor muscles and the elastic energy storage in the Achilles tendon. Although plantar flexor muscle mechanics and Achilles tendon energy storage have been explored during rearfoot striking, they have not been fully characterized during forefoot striking. This study examines how plantar flexor muscle-tendon mechanics during running differ between rearfoot and forefoot striking. We used musculoskeletal simulations, driven by joint angles and electromyography recorded from runners using both rearfoot and forefoot striking running patterns, to characterize plantar flexor muscle-tendon mechanics. The simulations revealed that foot strike pattern affected the soleus and gastrocnemius differently. For the soleus, forefoot striking resulted in decreased tendon energy storage and decreased positive fiber work compared to rearfoot striking. For the gastrocnemius, forefoot striking resulted in greater activation and increased negative fiber work compared to rearfoot striking. The increased activation and negative fiber work in the gastrocnemius during forefoot striking suggest that runners planning to convert to forefoot striking would benefit from a progressive eccentric gastrocnemius strengthening program to avoid injury.

bioengineering↗

bioRxiv2019

Algorithmic differentiation improves the computational efficiency of OpenSim-based optimal control simulations of movement

Algorithmic differentiation (AD) is an alternative to finite differences (FD) for evaluating function derivatives. The primarily aim of this study was to demonstrate the computational benefits of using AD instead of FD in OpenSim-based optimal control simulations. The secondary aim was to evaluate computational choices including different AD tools, different linear solvers, and the use of first- or second-order derivatives. First, we enabled the use of AD in OpenSim through a custom source code transformation tool and through the operator overloading tool ADOL-C. Second, we developed an interface between OpenSim and CasADi to perform optimal control simulations. Third, we evaluated computational choices through simulations of perturbed balance, two-dimensional predictive simulations of walking, and three-dimensional tracking simulations of walking. We performed all simulations using direct collocation and implicit differential equations. Using AD through our custom tool was between 1.8 {+/-} 0.1 and 17.8 {+/-} 4.9 times faster than using FD, and between 3.6 {+/-} 0.3 and 12.3 {+/-} 1.3 times faster than using AD through ADOL-C. The linear solver efficiency was problem-dependent and no solver was consistently more efficient. Using second-order derivatives was more efficient for balance simulations but less efficient for walking simulations. The walking simulations were physiologically realistic. These results highlight how the use of AD drastically decreases computational time of optimal control simulations as compared to more common FD. Overall, combining AD with direct collocation and implicit differential equations decreases the computational burden of optimal control simulations, which will facilitate their use for biomechanical applications.

bioengineering↗