EMG Signal Acquisition and Processing for Muscle Contraction Classification

Nowadays, the focus is on portable, non-invasive devices with a variety of functions, such as continuous monitoring through smartwatches or biologic signal-controlled prosthetic limb control. The signal acquisition and processing steps are the underlying principles behind all of these applications. This study proposes a low-budget Arduino-based acquisition system for recording forearm surface electromyography signals, which are then analyzed using Matlab software. The signals are recorded from four right-handed healthy individuals, following an established protocol consisting of ten 30 -second recordings per hand, with movements that alternate between five seconds of contraction and five seconds of relaxation. The classification results are then achieved by processing the data using the Support Vector Machine algorithm and GridSearch. There are four experiments conducted. The first one distinguishes between contraction and relaxation, then three additional experiments are performed following the same procedure (discriminations between gender, hand, gender and hand). Subsequently, a graphical user interface is designed to display the uploaded signal together with some details about the patients, the experiment, and the classification outcomes. The results achieved with the Arduino microcontroller are confirmed by performing the same steps using the Biopac system and comparing the outcomes in terms of metrics.