Abstract
Thermal imaging is progressively being used nowadays to investigate a wide range of muscle disorders or irregularities causing increased or decreased blood flow. Furthermore, the activation of the muscle tissue can accurately be detected using electromyography technique by measuring the electrical activity produced during muscular contraction and relaxation. In combination with smart material-based actuators and sensors, which are versatile and appropriate for integration, a multisensor feedback unit for human interaction can be developed. This research presents a novel approach for investigating muscle activation level and localization using surface temperature based on low-cost surface electromyography and thermal sensor that incorporates multi-mode dielectric elastomer actuator feedback capability. We have implemented a deep learning algorithm to make prediction on muscle activation and localization for the purpose of optimization. A graphical user interface is developed to make predictions and forecast on the go. The deep learning model exhibits high evaluation scores across accuracy metrics with an accuracy of 92% on the test set. The study explores the potential applications of this approach in fields of biomechanics, rehabilitation, and human-machine interfaces.
