Enhancing mobile EEG: Software development and performance insights of the DreamMachine

Abstract

Electroencephalography (EEG) is widely used in fields such as neurology, cognitive neuroscience, sleep research, and mental health. It records brain electrical activity to study neurophysiological functions. Numerous EEG and mobile EEG systems are available. However, adherence to the standards set by the International Federation of Clinical Neurophysiology (IFCN) is essential for ensuring high-quality data collection in clinical environments. The DreamMachine, a mobile EEG device, complies fully with these standards, offering 24-channel recordings at 250 Hz, Bluetooth Low Energy (BLE), and capabilities for electrooculography (EOG) and electrocardiography (ECG). Its low cost makes it an accessible option for EEG studies. The software architecture of the open-source DreamMachine is detailed in this study. Focus is placed on data compression and communication between the device and its companion Android application. The details of the Android application’s features, including gain settings, bits per channel, filters, bit-shifting, and safety factors, are investigated. Subsequently, the system’s performance is evaluated through a standard eyes-open/eyes-closed experiment, comparing its results with a laboratory EEG system across a significant number of participants to assess the performance of the DreamMachine system.