NIR-IIb Fluorescence Molecular Tomography of Glioblastomas Based on Heterogeneous Mouse Models and Adaptive Projection Match Pursuit Method.

Abstract

Fluorescence molecular tomography (FMT) using the second near-infrared window (NIR-II) fluorescence has been proved to outperform conventional FMT using the first near-infrared window (NIR-I) fluorescence. However, it was still a challenge to achieve a satisfactory reconstructed light source using NIR-II FMT as the NIR-IIa (1300-1400 nm) fluorescence in the NIR-II spectrum used in the previous NIR-II FMT study was still suffering from prominent absorption and scattering of tissue. A novel NIR-IIb (1500-1700 nm) FMT method was proposed and applied in the reconstruction of glioblastomas in animal models. Optical parameters that describe the effect of different tissue on the NIR-IIb photons were calculated to construct a light propagation model of NIR-IIb light to form the forward model. Besides, a novel adaptive projection matching pursuit (APMP) method was further adopted to accurately solve the inverse problem. Location error and Dice coefficient were used to evaluate the accuracy of reconstruction. Simulation experiments using single-source and dual-source and in vivo experiments were conducted to evaluate the reconstructed light source. Both simulation and in vivo experiments demonstrated that NIR-IIb FMT achieved reduced location error and improved shape similarity compared with NIR-IIa. NIR-IIb FMT of orthotopic mouse models further achieved a location error of 0.65 mm and a Dice coefficient of 0.56, which both outperformed NIR-IIa. The results demonstrated that NIR-IIb has better reconstruction performance for positioning accuracy and shape recovery. The inspiring results in this study demonstrate the effectiveness and advantages of NIR-IIb FMT in precise tumor positioning.