Three dimensional simulation of functional neuro-vascular imaging

Functional optical imaging has become a powerful tool for measuring physiological parameters in the brain without disrupting normal physiology. Fluorescence lifetime imaging (FLIM) has been shown to allow near real time mapping of oxygen tension in plasma (pO2), and Laser Speckle Contrast Imaging (LSCI) has been demonstrated to provide qualitative assessments of blood flow in the cortex. However, as both of these methods provide physiological parameters based on the spatial sampling of photons arriving at a detector, it is crucially important to understand either where the photons originated, in the case of FLIM, or which moving particles the photons have sampled, in the case of LSCI. Traditionally, these questions have been difficult to solve because of the heterogeneity of the distribution of particles which contribute to the measured signal. In both FLIM and LSCI, for example, only the light which samples the intravascular space will contribute to the signal. While analytical methods have proven to be successful at predicting the imaging depth of homogeneous materials, they are not able to predict imaging depth when measuring a fluorophore or a moving particle that is only present inside blood vessels. Unlike analytical methods, numerical methods can be used to approximate light propagation in an arbitrary geometry. While both deterministic and stochastic models of light propagation can, and have been, successfully employed to determine light fluence in an arbitrary geometry, deterministic methods are not well suited to the task of simulating light propagation in large volumes of turbid media. For this reason, three dimensional Monte Carlo simulations of light propagation combined with high resolution vascular anatomy were used to directly simulate FLIM and LSCI in the brain. Using these simulations, the imaging depth, degree of multiple scattering, and sensitivity of LSCI and FLIM to physiological changes were determined.