FastEMD–CCA algorithm for unsupervised and fast removal of eyeblink artifacts from electroencephalogram

•The algorithm yields very high SNR, denoting higher ratio of neural information preservation.•The effectiveness of the evaluated algorithm in preserving the underlying neural information in an EEG signal is proved through CC value that approaches near 1, low RMSE and higher SNR value.•The computation time of the algorithm is very low, with an average of 63 ms processing time to remove artifacts from 1 s length of EEG signal with 14 channels (256 samples × 14 EEG channels).•The algorithm is a feasible solution for applications requiring online removal of eyeblink artifacts, with typically low distortion to the neural signal.