FAU researchers create new model for developing assistive systems
Immense progress has been made in recent years in developing assistive systems for people who are paralyzed. However, the underlying neuromotor processes – from the brain’s commands through the neural circuits in the spinal cord to the activation of the muscles – have so far been mostly studied in isolation. Researchers at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) and their partners in Brazil have developed the MyoGen platform, which integrates these processes into a single simulation. The model, presented in the journal Nature Communications*, is freely accessible and is intended to support both clinical research and the optimization of medical devices.
Capturing and interpreting electrical signals from muscle activity is the key to developing wearable assistive systems. “Spinal cord injuries often lead to paralysis of the limbs because important signaling pathways from the brain to the muscles no longer function,” says Raul Sîmpetru of the Neuromuscular Physiology and Neural Interfacing Laboratory (N-squared Lab) at FAU. “However, the neural connections are not always completely interrupted, which is why residual muscle activity can be measured with electrodes.” During his doctoral research, Sîmpetru investigated how electromyography can be used for AI-driven assistive systems. The N-squared Lab develops such devices, including orthoses that help people with impaired hand function grasp objects.
Despite significant progress, the fundamental mechanisms of neuromotor control are still not fully understood. Raul Sîmpetru of FAU and Ricardo G. Molinari of Universidade Estadual de Campinas (Unicamp) in Brazil, pooled their expertise in data modeling and experimental research to develop the MyoGen platform. MyoGen incorporates motor commands from the brain, neural circuits in the spinal cord, feedback from muscles and tendons, and the generation of muscle force and electrical muscle signals into one integrated simulation. The platform draws on published physiological models and experimental findings on nerve cells, muscles, and tendons. “Many of these processes are well studied but have so far mostly been considered in isolation. We have brought them together in one model,” explains Sîmpetru. The Universidade Federal do ABC in Brazil is also involved in the project, alongside the N-squared Lab and Unicamp.
Reference model for research and development
The simulation shows exactly which nerve cells activate the muscle fibers and when. In this way, MyoGen provides reference data for both clinical research and the development of commercial assistive systems: “Many processes that affect how nerves and muscles work together cannot be directly observed or measured independently of one another in humans,” says Ricardo G. Molinari. “MyoGen allows us to investigate, under controlled conditions, how nerve signals activate muscles, generating force and electrical signals. This helps us develop and thoroughly test new methods for analyzing muscle signals.”
The platform, which is continuously expanded and updated, is designed as an open-source model and is therefore freely accessible for research and development. The modular design makes it possible to adapt individual components as needed and connect them. Intermediate results that have already been calculated can be reused as long as they are unaffected by any modifications. The MyoGen team is counting on feedback from the scientific community: Suggestions for extensions and improvements will be taken into account in future development.
Further information:
“MyoGen: Unified Biophysical Modeling of Human Neuromotor Activity and Resulting Signals”
*DOI: 10.1038/s41467-026-79273-3
