hmrR_MotionCorrectSplineSG - BUNPC/Homer3 GitHub Wiki

UI Name :
SplineSG_Motion_Correction

Syntax :
data_d = hmrR_MotionCorrectSplineSG(data_d, mlActAuto, p, FrameSize_sec, turnon)

Description :
The function first identifies the baseline shifts and spikes in the signal. The baseline shifts are corrected using a spline interpolation method. The remaining spikes are corrected by Savitzky-Golay filtering, which is a digital smoothing filter that substitutes each value of the signal series with a new value obtained by applying a cubic fitting to a subset of adjacent data points. The length of the subset is defined by the parameter FrameSize_sec. The method is described in the following paper. Citation: Jahani et al., Neurophotonics, 2018; doi: 10.1117/1.NPh.5.1.015003.

Inputs :

• data_d: SNIRF object containing time course data.
• mlActAuto:
• p: Parameter p used in the spline interpolation. The value recommended in the literature is 0.99. Use -1 if you want to skip this motion correction.
• FrameSize_sec:
• turnon: Optional argument to enable/disable this function in a processing stream chain.

Outputs :

• data_d: SNIRF object containing time course data after spline interpolation correction.

Usage Options :

• SplineSG_Motion_Correction: dod = hmrR_MotionCorrectSplineSG(dod, mlActAuto, p, FrameSize_sec, turnon)

Default Parameters :

• p=0.99
• FrameSize_sec=10
• turnon=1

Prerequisites :

• Intensity_to_Delta_OD: dod = hmrR_Intensity2OD( intensity )