Predictive Framework for Electrical Stimulation Cycling in Spinal Cord Injury

Enhancing the efficacy of spinal cord injury (SCI) rehabilitation is crucial for a patient's optimal recovery. While functional electrical stimulation (FES) cycling stands as a standard therapy, achieving notable improvements proves challenging due to the inherent complexities embedded in the dynamics of the movement. Indeed, overcoming the time-consuming parameter selection process becomes imperative, prompting the development of predictive models through optimal control simulation. The current challenge lies in the demand for a blueprint that considers the unique particularities of SCI FES cycling. In response, our innovative approach introduces a novel framework and showcases its application in solving predictive models. Leveraging open-source tools, including OpenSim and Blender, we built the FES cycling model. Subsequently, we outlined two problems formulations within OpenSim Moco: (P1) moving from point A to point B with different crank resistances, and (P2) tracking target speeds. Our study reveals the successful convergence of these simulations, demonstrating the integrated framework's robustness and efficiency. Indeed, the presented solution addresses the need for multiple simulations, thereby mitigating the lengthy constraints of prior methods and paving the way for practical and time-effective integration of digital twins in clinical applications.