Single-View Fluorescence Molecular Tomography Based on Hyperspectral NIR-II Imaging

Biological tissue optics has garnered significant attention in biomedical research due to its non-destructive nature and high sensitivity. However, the scattering and absorption properties of biological tissues fundamentally limit the penetration depth of optical imaging techniques. Fluorescence molecular tomography (FMT) has emerged as a promising solution that balances imaging depth and resolution, yet tissue scattering and absorption continue to challenge depth-resolved reconstruction accuracy. This study develops a highly sensitive near-infrared II (NIR-II) hyperspectral imaging system and systematically investigates, both experimentally and theoretically, the relationship between fluorescence penetration depth and tissue absorption/scattering coefficients. By leveraging the strong water absorption peak around 1450 nm, which results in shallow penetration depth, we strategically divide the reconstruction object into natural layers within the FMT model, thereby significantly improving the ill-posed inverse problem. Furthermore, we utilize the high-throughput continuous spectral data from hyperspectral imaging to select a series of wavelengths with progressively decreasing absorption coefficients relative to the 1450 nm water absorption peak. This enables layer-by-layer three-dimensional reconstruction of deep biological tissues, overcoming the fundamental limitations of conventional FMT approaches. Through the above methods, we demonstrate singleperspective FMT reconstruction capable of resolving heterogeneous targets at 10 mm depth with exceptional 0.74 Dice coefficient accuracy in depth discrimination. This study presents a spectral-dimension-enhanced FMT method that enables accurate 3D reconstruction from single-view measurements. By systematically exploiting the depth-dependent light-tissue interactions at selected NIR-II wavelengths, our approach achieves comparable imaging quality to conventional multi-angle systems while significantly simplifying the experimental setup. Both simulation and phantom experiments demonstrate the method's capability for precise target localization and shape recovery, suggesting promising potential for practical small animal imaging applications where system complexity and acquisition speed are critical considerations.