A Mechanomyography-Based System for Application and Tuning of Transcutaneous Spinal Cord Stimulation

For patients with lesions in the upper motor neurons, transcutaneous spinal cord stimulation is a promising therapy technique to reduce spasticity and enhance voluntary movement. However, identifying an individual suitable therapy intensity and electrode position requires a time-consuming tuning procedure. This process typically entails the interpretation of electromyographic responses of leg muscles to double stimulation pulses to ensure the activation of targeted structures. To simplify and automate this process, we introduce a system consisting of a stimulator, a tablet, and four inertial measurement units used for mechanomyography (MMG). A pre-trained machine learning model is applied to the MMG signal and employs the response characteristics. The system derives a suitable therapy intensity for the selected electrode position from the model's results and assesses the suitability of this position with a score. Transferring classic EMG measurements to MMG-based systems and the automation of intensity identification might ease application and reduce time constraints in clinical settings.