FMRI denoising pipeline
Alfonso Nieto-Castañón
Abstract
After the functional data has been preprocessed, the measured blood-oxygen-level-dependent (BOLD) signal often still contains a considerable amount of noise from a combination of physiological effects, outliers, and residual subject-motion factors. If unaccounted for, these factors would introduce very strong and noticeable biases in all functional connectivity measures. This chapter describes standard and advanced denoising procedures in CONN that are used to characterize and remove the effect of these residual non-neural noise sources.
Topics & Concepts
Noise reductionResidualOutlierNoise (video)Pipeline (software)Computer scienceArtificial intelligenceSIGNAL (programming language)Pattern recognition (psychology)AlgorithmProgramming languageImage (mathematics)Non-Invasive Vital Sign Monitoring