Improving quantifying tissue optical properties of human heads in vivo by combining CW-NIRS and TR-NIRS (Conference Presentation)

Using near-infrared light to quantify optical properties (OPs) of the human head from the superficial to the brain could help using functional near-infrared spectroscopy (fNIRS) techniques to quantify the brain’s functional activity and optimizing transcranial stimulation by predicting photon energy distribution. We combine continuous-wave NIRS (CW-NIRS), time-resolved (TR) NIRS, and three-dimensional (3D) Monte Carlo (MC) simulations accelerated by artificial neural networks (ANN) to improve the accuracy in quantifying the scattering and absorption coefficients of the gray matter in vivo. Theoretical evaluation showed root-mean-square errors (RMSPE) of estimated OPs below 10% except for the scattering coefficient of gray matter, whose RMSPE was 15%. By a new detection optode design incorporating a right-angle prism, the fluctuations in repeated CW-NIRS measurements were reduced by 34%.