A new method to estimate abundances of multiple components using multi-spectral fluorescence lifetime imaging...

Area of Science:

• Biomedical Optics
• Medical Imaging
• Molecular Spectroscopy

Background:

• Multi-Spectral Fluorescent Lifetime Imaging Microscopy (m-FLIM) measures endogenous fluorescence decay for noninvasive diagnosis.
• Tissue fluorescence signals are complex mixtures of multiple molecular components.
• Accurate decomposition of these mixtures is crucial for molecular characterization.

Purpose of the Study:

• To develop a novel method for estimating the abundances of multiple fluorescent components in m-FLIM data.
• To provide a quantitative description of molecular constitution for improved in situ clinical diagnosis.

Main Methods:

• A closed-form solution for linear unmixing is proposed under the fully constrained model.
• Assumes known number of components and their ideal lifetime decays.
• Validated using synthetic samples and ex vivo human atherosclerotic tissue.

Main Results:

• The method achieved approximately 6% error on synthetic samples with three components.
• Demonstrated ~88% accuracy in unmixing ex vivo atherosclerotic tissue (collagen, elastin, LDL).
• Quantitative description of molecular composition was validated against ground-truth maps.

Conclusions:

• The developed method offers accurate and efficient unmixing of m-FLIM data.
• This technique has significant potential for noninvasive disease diagnosis and molecular profiling.
• Validated performance in complex biological samples highlights its clinical applicability.