Algorithmic depth compensation improves quantification and noise suppression in functional diffuse optical tomography

Accurate depth localization and quantitative recovery of a regional activation are the major challenges in functional diffuse optical tomography (DOT). The photon density drops severely with increased depth, for which conventional DOT reconstruction yields poor depth localization and quantitative recovery. Recently we have developed a depth compensation algorithm (DCA) to improve the depth localization in DOT. In this paper, we present an approach based on the depth-compensated reconstruction to improve the quantification in DOT by forming a spatial prior. Simulative experiments are conducted to demonstrate the usefulness of this approach. Moreover, noise suppression is a key to success in DOT which also affects the depth localization and quantification. We present quantitative analysis and comparison on noise suppression in DOT with and without depth compensation. The study reveals that appropriate combination of depth-compensated reconstruction with the spatial prior can provide accurate depth localization and improved quantification at variable noise levels.