Table 9.

Signal processing and artifact removal methods for fNIRS studies.

Name, year	Pre-processing	Removal of movement artifact
Al-Yahya et al., 2016	Low-pass filter at 0.67 Hz cutoff frequency
Beurskens et al., 2014	Gaussian/Hemodynamic response function lowpass filter	Wavelet-minimum description length algorithm
Caliandro et al., 2015	Low pass filter (0.1 Hz)
Doi et al., 2013	Low pass filter (0.5 Hz)
Fujita et al., 2016	Low pass filter (0.5 Hz)
	High pass filter (0.01 Hz)
Holtzer et al., 2011	Low pass filter (FIR, 0.14 Hz)	ICA and PCA
Holtzer et al., 2015, 2016	Low pass filter (FIR, 0.14 Hz)	Inspection to remove signal artifact
Huppert et al., 2013	Low pass filter (0.8 Hz)
	Series of discrete cosine transform terms
Karim et al., 2013	Series of discrete cosine transform terms
Kim et al., 2016		Gaussian smoothing. Wavelet minimum description length algorithm.
Koenraadt et al., 2014	Low pass filter (Butterworth, 1.25 Hz)	Short separation channels and scaling factor used to normalize data per individual
	High pass filter (Butterworth, 0.01 Hz)
	Low pass filter (Butterworth, 1 Hz)
Kurz et al., 2012	High pass filter (0.01 Hz)	PCA, removing components <0.25 correlation with reference waveform
Lin and Lin, 2016	Low pass filter (FIR, 0.2 Hz)
Lu et al., 2015	Removal of noisy channels using coefficient of variation Bandpass filter (0.01–0.2 Hz)	PCA and Spike Rejection
Maidan et al., 2015	Low pass filter (FIR, 0.14 Hz)
Mihara et al., 2008	High pass filter (0.05 Hz)	Gaussian function
Mihara et al., 2012	High pass filter (0.03 Hz)
Mirelman et al., 2014	Low pass filter (FIR, 0.14 Hz)
Takeuchi et al., 2016	Bandpass filter (0.01–0.5 Hz)	Rapid changes in oxyHb concentration were removed

FIR, finite impulse response; ICA, independent component analysis; PCA, principle component analysis; SD, standard deviation.