An Automatic Motion-Based Artifact Reduction Algorithm for fNIRS in Concurrent Functional Magnetic Resonance Imaging Studies (AMARA–fMRI)

Multimodal functional near-infrared spectroscopy–functional magnetic resonance imaging (fNIRS–fMRI) studies have been highly beneficial for both the fNIRS and fMRI field as, example, they shed light on underlying mechanism of each method. However, several noise sources exist in methods. Motion artifact removal is an important preprocessing step analysis. Several manual motion–artifact methods developed which require time are dependent expertise. Only a few automatic proposed. AMARA (acceleration-based movement reduction algorithm) one most promising was originally tested sleep study with long acquisition times (~8 h). it relies accelerometry data, problematic when performing concurrent fNIRS–fMIRI experiments. Most accelerometers not MR compatible, any case, existing datasets do this data. Here, we propose new way to retrospectively determine acceleration data motion correction methods, such as multimodal fNIRS–fMRI studies. We so by considering individual slice stack simultaneous multislice (SMS) reconstructing high-resolution traces from time. validated our method 10 participants during memory task (2- 3-back) 6 channels over prefrontal cortex (limited view fMRI). found that significantly improved detection activation deoxyhemoglobin outperformed up-sampled traces. no improvement oxyhemoglobin. Furthermore, show high overlap according