Macroscopic fluorescence lifetime topography enhanced via spatial frequency domain imaging

Area of Science:

• Biomedical Optics
• Medical Imaging
• Computational Biology

Background:

• Accurate depth localization of fluorescent inclusions in bio-tissues is crucial for applications like optical-guided surgery.
• Existing methods may face challenges in precisely determining the depth of deeply seated fluorescent targets.

Purpose of the Study:

• To develop a computational framework for macroscopic fluorescence lifetime imaging (MFLI) topography.
• To accurately retrieve the depth of fluorescent inclusions within bio-tissues using machine learning.

Main Methods:

• A machine learning-based computational framework utilizing macroscopic fluorescence lifetime imaging (MFLI) topography.
• Integration of spatial frequency domain imaging (SFDI) for in situ optical property retrieval.
• A Siamese network architecture processing optical properties and time-resolved fluorescence decays to simultaneously determine lifetime maps and depth profiles.

Main Results:

• The developed MFLI topography framework accurately retrieves the depth of fluorescent inclusions.
• Coupling depth retrieval with optical property estimation significantly enhances accuracy.
• Validation performed using extensive in silico datasets and phantom experiments confirmed the approach's efficacy.

Conclusions:

• The proposed machine learning-based MFLI topography framework offers a highly accurate method for depth retrieval of fluorescent inclusions.
• This computational approach is expected to be valuable for advancing optical-guided surgery and other biomedical imaging applications.