Short-channel regression in functional near-infrared spectroscopy is more effective when considering heterogeneous scalp hemodynamics

Abstract

. Significance: The reliability of functional near-infrared spectroscopy (fNIRS) measurements is reduced by systemic physiology. Short-channel regression algorithms aim at removing systemic “noise” by subtracting the signal measured at a short source–detector separation (mainly scalp hemodynamics) from the one of a long separation (brain and scalp hemodynamics). In literature, incongruent approaches on the selection of the optimal regressor signal are reported based on different assumptions on scalp hemodynamics properties. Aim: We investigated the spatial and temporal distribution of scalp hemodynamics over the sensorimotor cortex and evaluated its influence on the effectiveness of short-channel regressions. Approach: We performed hand-grasping and resting-state experiments with five subjects, measuring with 16 optodes over sensorimotor areas, including eight 8-mm channels. We performed detailed correlation analyses of scalp hemodynamics and evaluated 180 hand-grasping and 270 simulated (overlaid on resting-state measurements) trials. Five short-channel regressor combinations were implemented with general linear models. Three were chosen according to literature, and two were proposed based on additional physiological assumptions [considering multiple short channels and their Mayer wave (MW) oscillations]. Results: We found heterogeneous hemodynamics in the scalp, coming on top of a global close-to-homogeneous behavior (correlation 0.69 to 0.92). The results further demonstrate that short-channel regression always improves brain activity estimates but that better results are obtained when heterogeneity is assumed. In particular, we highlight that short-channel regression is more effective when combining multiple scalp regressors and when MWs are additionally included. Conclusion: We shed light on the selection of optimal regressor signals for improving the removal of systemic physiological artifacts in fNIRS. We conclude that short-channel regression is most effective when assuming heterogeneous hemodynamics, in particular when combining spatial- and frequency-specific information. A better understanding of scalp hemodynamics and more effective short-channel regression will promote more accurate assessments of functional brain activity in clinical and research settings.

Comparison of short-channel separation and spatial domain filtering for removal of non-neural components in functional near-infrared spectroscopy signals

“…Others have argued that it is necessary to record short-channel information as well as additional sources of physiology from multiple regions of the scalp to determine local influences of scalp hemodynamics and additionally remove them. 24 , 49 , 50 Another recent method for removing systemic information has been proposed that utilizes multi-distance tomographic recordings to separate superficial from cortical hemodynamics. 41 The effectiveness of short-channel regression has been further investigated and compared to recently developed methods that are also intended to separate systemic from cortical responses in fNIRS recordings.…”

Cortical hemodynamic mapping of subthalamic nucleus deep brain stimulation in Parkinsonian patients, using high-density functional near-infrared spectroscopy

“…While the systemic physiological interference distribute across all the channels over the head, our result showed relatively local spatial distribution of HbO and HbR concentration values mapped over the motor/premotor and frontal dorsolateral cortices ( Fig 2B ) using high-resolution optode distributed over most of the head ( Fig 1B ), however, we cannot completely exclude the influence of these confounding factors. Hence, one limitation of current study is that influence of DBS-induced decreases in skin blood flow cannot definitely be denied since changes in systemic physiological factors (e.g., cardiac, respiratory, and blood pressure fluctuations) can affect regional blood flow and some studies have revealed that skin blood flow changes, which can impact changes in HbO, are not homogeneous [ 147 – 149 ]. Therefore, the use of additional short-distance channels (<1 cm) is expected to provide a superior removal of the superficial systemic factors on functional near-infrared spectroscopy [ 149 , 150 ].…”

“…Hence, one limitation of current study is that influence of DBS-induced decreases in skin blood flow cannot definitely be denied since changes in systemic physiological factors (e.g., cardiac, respiratory, and blood pressure fluctuations) can affect regional blood flow and some studies have revealed that skin blood flow changes, which can impact changes in HbO, are not homogeneous [ 147 – 149 ]. Therefore, the use of additional short-distance channels (<1 cm) is expected to provide a superior removal of the superficial systemic factors on functional near-infrared spectroscopy [ 149 , 150 ].…”

Amplitude of fNIRS Resting-State Global Signal Is Related to EEG Vigilance Measures: A Simultaneous fNIRS and EEG Study

“…Later, most resting-state fMRI studies have adopted GSR as a pre-processing approach: the global signal component is regressed out of preprocessed BOLD signals prior to computation of connectivity measures and therefore regionally focused connectivity patterns are reported (Fox et al, 2009). Similarly, in recent fNIRS studies of resting state brain, a global component has been recognized in the measurements from regularly distanced optodes (White et al, 2009;Mesquita et al, 2010;Tong and Frederick, 2010;Eggebrecht et al, 2014;Tachtsidis and Scholkmann, 2016;Duan et al, 2018;Wyser et al, 2020) and from short-distanced optodes (White et al, 2009;Gregg et al, 2010;Mesquita et al, 2010;Eggebrecht et al, 2014;Tachtsidis and Scholkmann, 2016;Duan et al, 2018;Sherafati et al, 2020;Wyser et al, 2020). To date, there is no wellestablished pre-processing routine in resting state fNIRS studies although multiple efforts are being made (Huppert et al, 2009;Ye et al, 2009;Xu et al, 2014;Santosa et al, 2018).…”

“…The most common RSFC analysis of fNIRS data involves evaluating the temporal relationship between time series of the preprocessed data from recording units, for example, through the Pearson’s correlation. A global component has been observed in fNIRS measurements and commonly removed for the purpose of attenuating systematic noises at the resting state ( White et al, 2009 ; Gregg et al, 2010 ; Mesquita et al, 2010 ; Eggebrecht et al, 2014 ; Tachtsidis and Scholkmann, 2016 ; Duan et al, 2018 ; Sherafati et al, 2020 ; Wyser et al, 2020 ). Whereas removing superficial contributions from short-distanced channels to fNIRS is increasingly employed to attenuate the systematic noises ( Saager and Berger, 2005 ; Gagnon et al, 2011 ), data from both long-distanced and short-distanced channels commonly suggest a global component exist in fNIRS measurements and distribute across wide regions ( Zhang et al, 2005 , 2007 , 2009 ; Kohno et al, 2007 ; Gregg et al, 2010 ; Tong and Frederick, 2010 ; Novi et al, 2016 ; Sato et al, 2016 ).…”