In-silico neuro-musculoskeletal model demonstrates spasticity progression with descending motor tracts loss in simulated clinical triage.

Our study explores the complex mechanisms of spasticity using a multi-scale neuro-musculoskeletal model to investigate its emergence and characteristics following spinal cord injury. We built a large-scale, biologically realistic, closed-loop spino-musculoskeletal model of the lower limb using the NEUROiD co-simulation platform. The in-silico spinal cord incorporated around 50 spinal pathways, 40,000 alpha motor neurons, and approximately 12 million interconnections spanning L2-S2 and was simulated on NEURON. The musculoskeletal model contained 92 muscles and was simulated on OpenSim. We conducted in-silico triage of two clinical assessments of spasticity, the Modified Tardieu Test and Pendulum Test, on this model and compared results with published literature. We also demonstrated subject-specific customized models (virtual subjects) using motion capture data. Our model demonstrated the clasp-knife effect in the Modified Tardieu Test simulations and damped knee oscillations with increased spinal injury in Pendulum Test simulations. We also demonstrated how spastic behavior emerges with varying degrees of supra-spinal disruption. Further, using data from motion capture experiments, we explore subject-specificity in our model. Our work presents methods for creating a virtual subject with triaging capabilities that can demonstrate the onset of spasticity symptoms and its possible subject-specific variances.