In-silico neuro musculoskeletal model reproduces the movement types obtained by spinal micro stimulation

•An in-silico upper limb model has been built using the NEUROID platform. The multiscale model has a modular architecture and facilitates the reuse of existing models in NEURON and Opensim, making it feasible to quickly curate and integrate models to build larger sub-systems and systems.•A notable feature in the spinal cord model is that all elements from ion channels to neurons are placed within anatomical coordinates. This opens the possibility of simulating anatomical interventions and their resultant movement behaviors.•By means of spinal cord stimulations and reminiscent of descending cortical activations, emergence of common upper limb movements were demonstrated.•The movements produced by stimulation of various spinal levels produced a map of various degrees of movement along the spinal cord. This map was in agreement with known experimental results. Virtual patients and physiologies allow experimentation, design, and early-stage clinical trials in-silico. Virtual patient technology for human movement systems that encompasses musculoskeleton and its neural control are few and far in between. Our major goal is to create a neuro- musculoskeletal upper limb in-silico model, which is modular in architecture and generates movement as an emergent phenomenon out of a multiscale co-simulation of spinal cord neural control and musculoskeletal dynamics. The model is developed on the NEUROiD movement simulation platform that enables a co-simulation of popular neural simulator NEURON and the musculoskeletal simulator OpenSim. We further characterized and demonstrated the use of this model in generating a range of commonly observed upper limb movements by means of a spatio-temporal stimulation pattern delivered to the cervical spinal cord. We were able to characterize the model based on proprioception (Ia, Ib and II fibers), afferent conduction delay and inital postures of the musculoskeletal system. A smooth movement was achieved in all the considered experiments. The generated movements in all degrees of freedom were reproduced in accordance with the previous experimental studies. In this work, design and development of the upper limb model was described in a modular fashion, while reusing existing models and modules. We believe this work enables a first and small step towards an in-silico paradigms for understanding upper limb movement, disease pathology, medication, and rehabilitation.