Subject-dependent classification for robust idle state detection using multi-modal neuroimaging and data-fusion techniques in BCI

作者：

Highlights：

• EEG–NIRS measurements can robustly detect the idle class.

• An activation function to reduce hemodynamic response delays in NIRS is proposed.

• A novel hybrid classification strategy combining EEG–NIRS classifiers is proposed.

• Superior performance by complementary information of multimodal recordings.

摘要

Highlights•EEG–NIRS measurements can robustly detect the idle class.•An activation function to reduce hemodynamic response delays in NIRS is proposed.•A novel hybrid classification strategy combining EEG–NIRS classifiers is proposed.•Superior performance by complementary information of multimodal recordings.

论文关键词：Hybrid brain–computer interfacing,Combined EEG–NIRS,Classifier combination,Subject-dependent classification

论文评审过程：Received 23 May 2014, Revised 30 December 2014, Accepted 8 March 2015, Available online 18 March 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.03.010