Autofluorescence has been a significant disadvantage when dealing with tomographic imaging of biological samples or tissue phantoms. Consequently, the accurate removal of autofluorescence signal has been a major concern in fluorescence tomography. Here we present a study on three-dimensional mapping and removal of autofluorescence from fluorescence molecular tomography (FMT) data, both for phantoms and small animal in vivo. The technique is based on the recording of tomographic data in multiple spectral regions with different excitation light and on the application of a linear unmixing algorithm for targeting multiple fluorescent probes. Two types of measurements are taken, one with the excitation being in the region of the maximum absorption of the targeted fluorophore and one in a region away from the maximum. The relative strengths of the different spectra are employed to calculate the signal to be removed from the tomographic reconstruction. Autofluorescence spectra are recorded using identical reflection geometry as during the FMT acquisitions allowing for the correct mapping of the autofluorescence signal. Results from phantoms exhibiting different background autofluorescence strengths are presented and discussed. In this work we have also studied in vivo fluorescent activity in mice, involving both subcutaneously implanted fluorescent phantoms and b10 transgenic mice.
© 2007 SPIEPDF Article