Machine Learning and Signal Processing for Neurotechnologies and Brain-Computer Interactions Out of the Lab

A brain-computer interface (BCI) provides a direct communication pathway between a human brain and an external device, with the goal to assist, augment, or repair human cognitive or sensory-motor functions. Despite the impressive expansion in recent years, the BCI systems described in the literature are not sufficiently mature for the daily use out of the laboratory. The performance of the current BCI systems would greatly deteriorate if brain signals were recorded in an unrestricted outside-the-lab environment with limited control on artifacts and external distractions. Moreover, the long set up and calibration time would be major burdens in long-term use of BCI. The use of portable EEG devices with a few dry electrodes would greatly reduce the set up time, however, with the expense of having noisier and less stable brain signals. Hybrid BCIs where brain signals all combined with other modalities to improve the control would enhance practicality of BCI in outside environments.

This Research Topic aims to present rigorous, significant, and impactful studies that use advanced signal processing and machine learning algorithms to address issues impairing the usability of BCI out of the lab.

Experimental studies as well computational or theoretical works are both welcome. We accept both original articles and review papers. This includes, but is not limited to, algorithms and techniques dealing with noisy and non-stationary brain signals, user’s motion artifacts, long calibration time, variability in properties of brain signals across individuals or tasks, real-time BCI interactions, BCI high mental workload and user fatigue, accelerating learning and improving performance in short-term and long-term, multimodal and hybrid BCIs, new medical and non-medical applications of BCI, new BCI paradigms and new portable BCI devices.