Discrimination method between deep and shallow components in NIRS signal

摘要

A method using multi-distance optodes and independent component analysis (ICA) has been proposed to discriminate between the components of deep- and shallow-tissue layers included in near-infrared spectroscopy (NIRS) signals. The method assumes that the partial optical path length of the deep layer linearly increases along with the increase of the source-detector (S-D) distance, whereas that of the shallow layer does not change. Reconstruction of signals is performed by the summation of independent components weighted by the deep or shallow contribution. We applied the method to NIRS signals measured on human heads with 15- and 30-mm S-D distances during several tasks. The shallow signals had a higher temporal correlation with the laser-Doppler flowmetry (LDF) signals and with the 5-mm S-D distance channel than the deep signals, which showed the validity of the method. We further investigated the effect of delay time parameter in ICA on the performance of the method and found that the parameter does not affect the performance very much.