Time-resolved early-photon scheme for high-resolution fluorescence molecular tomography with perturbation Monte Carlo modeling: Experimental validations using a multichannel TCSPC system

Abstract

In near-infrared fluorescence molecular tomography (FMT), high turbidity of biological tissue exerts severe adversity to image reconstruction. In contrast to highly diffusive photons, early-photons preserve a significant amount of spatial information, and thereby considerably improve image quality by reducing ill-posedness in inversion processes. However, light propagation of early-photons is hard to be accurately modeled by the mathematically simple diffusion equation (DE), and the rich information in the time-resolved (TR) measurements has not been effectively utilized in reconstruction with the traditional single-gate (SG) data type. In order to take an advantage of the early-photons in time-domain FMT for in-vivo imaging of small animals, we have proposed an early-photon FMT based on a perturbation Monte-Carlo (pMC) modeling and an overlap-delaying TR data type. In this method, the fluorescence Jacobians are generated by the pMC-based forward calculations with calibration by measured instrumental response function for each source-detector pair, and the measured time-of-flight distributions of excitation and emission light are averaged over time-gates for enhancing the noise robustness. The proposed early-photon FMT based on TR-pMC has been verified by phantom experiments, demonstrating improved image qualities, such as the spatial resolution, gray-scale resolution, quantitativeness, and peak signal-to-noise ratio when compared with the early-photon FMT based on TR-DE and SG-pMC.