Reduction of global interference of scalp-hemodynamics in functional near-infrared spectroscopy using short distance probes

Abstract

Functional near-infrared spectroscopy (fNIRS) is used to measure cerebral activity because it is simple and portable. However, scalp-hemodynamics often contaminates fNIRS signals, leading to detection of cortical activity in regions that are actually inactive. Methods for removing these artifacts using standard source-detector distance channels (Long-channel) tend to over-estimate the artifacts, while methods using additional short source-detector distance channels (Short-channel) require numerous probes to cover broad cortical areas, which leads to a high cost and prolonged experimental time. Here, we propose a new method that effectively combines the existing techniques, preserving the accuracy of estimating cerebral activity and avoiding the disadvantages inherent when applying the techniques individually. Our new method accomplishes this by estimating a global scalp-hemodynamic component from a small number of Short-channels, and removing its influence from the Long-channels using a general linear model (GLM). To demonstrate the feasibility of this method, we collected fNIRS and functional magnetic resonance imaging (fMRI) measurements during a motor task. First, we measured changes in oxygenated hemoglobin concentration (∆Oxy-Hb) from 18 Short-channels placed over motor-related areas, and confirmed that the majority of scalp-hemodynamics was globally consistent and could be estimated from as few as four Short-channels using principal component analysis. We then measured ∆Oxy-Hb from 4 Short- and 43 Long-channels. The GLM identified cerebral activity comparable to that measured separately by fMRI, even when scalp-hemodynamics exhibited substantial task-related modulation. These results suggest that combining measurements from four Short-channels with a GLM provides robust estimation of cerebral activity at a low cost.

Detecting Motor Learning-Related fNIRS Activity by Applying Removal of Systemic Interferences

“…Here we showed that our method using just a few short distance probe channels (Short-channels) [9] can be used for scalp artifact removal during motor learning tasks. Nevertheless, several different artifact removal methods have been proposed so far [2].…”

“…In cases applying the removal we used a method to incorporate scalp hemodynamic components into the model, referred to as ShortPCA [9]. In this method, the equation used for the analysis of fNIRS Long-channel in each trial is as follows:…”

“…This was created by convoluting a 20-s boxcar function representing the rotation task with the canonical hemodynamic response function [12]. X s is a component (1 × T ) to model scalp hemodynamic artifacts and was estimated using principal component analysis (PCA) applied to Short-channel signals [9]. To remove the global scalp hemodynamic artifacts, which are distributed across channels, we considered three principal components as scalp hemodynamic artifacts because these could represent the majority of the Shortchannels (explained more than 90% of variance).…”

“…Such artifact is unavoidable since these changes occur on the surface (scalp) layer or even in the cortical layer, through which near-infrared light passes. Therefore, proper removal of scalp hemodynamic artifacts is critical for estimating true brain activity using fNIRS analysis [2], [9]. However, the effects of these artifacts during motor learning tasks have not been thoroughly investigated.…”

“…Thus, the first hypothesis proposes that if scalp hemodynamic artifacts constantly contaminate fNIRS signals across learning, no influences of the scalp artifacts on learning-related cerebral activity would To examine if scalp artifact removal is required for the detection of learning-related cerebral activity (i.e., which hypothesis is true), we measured brain activity by fNIRS while subjects performed a visuomotor tracking task [10]. As a scalp artifact removal method, we applied a method using short probe distance channels, which is proven to be effective for motor tasks in the previous study [9] and identified learning-related brain activity using a general linear model (GLM) with or without scalp hemodynamic artifact removal. By comparing the results with and without scalp artifact removal, we investigated if the detection of cerebral activity associated with motor learning is improved by the removal of scalp artifacts.…”