No abstract

No abstract

26Functional near infrared spectroscopy (fNIRS) is a neuroimaging technique that uses light 27 at two different wavelengths in the near infrared spectrum to estimate cerebral hemodynamic 28 response, based on concentration changes in both oxygenated and deoxygenated hemoglobin. A 29 multi-dimensional decomposition technique, parallel factor analysis (PARAFAC), has recently 30 been validated for the identification of artifacts and cerebral activation patterns in 31 electroencephalography. Our objective was to introduce and validate the use of the PARAFAC 32 model for the analysis of fNIRS data, which is inherently multidimensional (time, space, 33 wavelength). Eighteen healthy adults underwent fNIRS acquisition during a semantic verbal 34 fluency task. Both the signal-to-noise ratio and Pearson's correlation were used to evaluate the 35 efficacy of PARAFAC for motion artifact correction. Temporal, spatial and hemodynamic 36 characteristics of the PARAFAC component allowed us to validate the identification of task-related 37 cerebral activation. Using PARAFAC, results revealed a significant improvement in data quality 38 after movement artifact correction. Compared to traditional two-dimensional approaches (e.g. 39 ICA, PCA), PARAFAC offered some advantages. Although PARAFAC revealed a slightly more 40 distributed functional network, temporal and spatial characteristics of the task-related brain 41 activation identified with PARAFAC mostly overlapped with commonly used approaches (e.g. 42 general linear model, global average). This study describes the first implementation of PARAFAC 43 in fNIRS and supports it as a promising data-driven alternative for multi-dimensional data analyses 44 in fNIRS. 45

No abstract