Open Source Electromyography EMG Project

In the open source electromyography project shared here, the author used this product to create a computer game control effect.

The prototype uses a main power board, which contains an Arduino Nano 33 BLE Sense, and a boost converter to power a AAA battery, providing 5 volts to the Arduino and a series of connected EMG amplifier units. The EMG sensor is based on the INA128 instrumentation amplifier and is very simple to configure. EMG samples as well as data from the IMU on the Nano 33 BLE Sense are passed via Bluetooth to a connected PC running the PsyLink software stack. This is based on Python, using the BLE-GATT library for BT communication, PynPut to handle PC input devices (to emit keyboard and mouse events) and tensorflow for the machine learning aspect. The idea is to use machine learning from EMG data to correlate specific user interface events (such as key presses) and, with some training, be able to play games on PC with only hand/arm gestures. IMU data is used to enhance this, but in this demonstration, this is not entirely clear.