Establishing Quantitative Measures of Quality of Functional Near Infrared Spectroscopy Data
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Functional Near-Infrared Spectroscopy (fNIRS) is an optical neuroimaging technique that can be used to examine and quantify tissue hemodynamics on the brain. fNIRS signals are contaminated by measurement noises and physiology-based systemic noises, such as a periodic pulsation of optical signals associated with the cardiac activity. Several approaches exist to filter out all sorts of noises and to remove channels with a low signal-to-noise ratio (SNR) that are deemed unreliable to estimate cortical hemodynamics. However, amongst the systemic noises which are undesirable for cerebral hemodynamics, strong cardiac pulsations usually indicate a good contact between the optical probe and the scalp. This thesis aims at evaluating the performance of physiology-based measures of quality of fNIRS data, namely 1) the Scalp Contact Index (SCI) and 2) the Peak Power (PP) of the spectrum, and understand how would they vary as a function of a range of pair of wavelengths, and for experiments conducted with different experimental setups such as 1) the source-detector distance, 2) the integration time of photodetectors and 3) the anatomical location on the head where signals are collected. So, while keeping other parameters constant, we are going to vary only one parameter at a time and collect the data and compute the SCI and PP for that data to compare its quality.