作者: Shireen Fathima , Sheela Kiran Kore
DOI: 10.3389/FNINS.2020.546656
关键词:
摘要: Electroencephalogram (EEG) is one of the common modalities monitoring mental activities. Owing to non-invasive availability this system, its applicability has seen remarkable developments beyond medical use-cases. One such use case brain-computer interfaces (BCI). Such systems require usage high resolution-based multi-channel EEG devices so that data collection spans multiple locations brain like occipital, frontal, temporal, and on. This results in huge (with sampling rates) with channels inherent artifacts. Several challenges exist analyzing nature, for instance, selecting optimal number or deciding what best features rely on achieving better performance. The selection these variables complicated requires a lot domain knowledge monitoring, which not feasible always. Hence, optimization serves be an easy access tool deriving parameters. Considerable efforts formulating issues as problem have been laid. As result, various multi-objective constrained functions developed BCI achieved reliable outcomes device control neuro-prosthetic arms, application control, gaming, paper makes attempt study techniques BCI. outcomes, challenges, major observations approaches are discussed detail.