Skip to main content
. 2019 Apr 26;13:141. doi: 10.3389/fnhum.2019.00141

Figure 1.

Figure 1

The EEGLAB toolbox methods are listed on the left (in gray) and include the two core functions of ASR and rASR that are called by the function clean_asr. Parts of the algorithm that were adapted using geometry-aware methods are marked by dashed sections. The toolbox consists of several preprocessing and correction functions that together comprise artifact correction for multi-channel EEG data. Some of the functions marked in square brackets are computationally expensive and profit from many data, but they are typically not used for online applications, we therefore refrained from using them herein. When using the toolbox, all functions are called once, clean_asr is then calling the calibration and subsequently the processing repeatedly with short segments of the uncorrected data. The procedure is the same for the online and offline use of ASR and rASR.