Artificial Intelligence-based Approach for Gait Pattern Identification Using Surface Electromyography (SEMG)

Internet of Things (IoT) is gaining significant development for various applications, including healthcare and medical field. The connectivity of sensor nodes using IoT enables the possibility of extracting more features from the data gathered. This information can be used to prepare real-time assistive technology and help in supervision, alert generation, training, and logging the activity for future purposes. In this work wearable Electromyography (EMG) based system is being presented to measure gait parameters in everyday life. EMG is implemented for recording the electrical activity of muscle tissue. Therefore, surface EMG (SEMG) is employed in most research works to get information related to muscle movements in health and posture recognition. We use gait prediction of a person using SEMG to monitor daily living activities, which includes climbing, walking, jogging, and jumping. The MyoWare muscle sensor is used for the experiment. The fetched data is processed using a deep neural network model to recognize the above-mentioned activities. The 98.44 percent accuracy is observed with the convolutional neural network model.