Computational 3D multispectral fluorescence lifetime microscopy

Area of Science:

• Biophotonics and Imaging Science
• Microscopy Technology
• Computational Imaging

Background:

• Traditional fluorescence microscopy faces limitations in simultaneously achieving high spatial resolution, spectral information, and fluorescence lifetime measurements.
• Volumetric imaging often requires trade-offs between acquisition speed, resolution, and the amount of information gathered.
• Non-invasive, fast, and comprehensive characterization of fluorescent biological samples is crucial for advancing research.

Purpose of the Study:

• To introduce a novel 3D multispectral fluorescence lifetime microscope integrating structured illumination microscopy (SIM) and single-pixel camera (SPC) techniques.
• To develop a computational framework for fusing high-resolution spatial data with spatial-temporal spectral data.
• To demonstrate the system's capability for fast, non-invasive, and high-fidelity volumetric fluorescence imaging.

Main Methods:

• Integration of SIM for high-resolution spatial data and SPC for spectral-temporal data acquisition.
• Development of data fusion algorithms to combine datasets into a five-dimensional representation.
• Application of compressive sensing to accelerate SPC data acquisition.
• Experimental validation using fluorescent beads and cellular samples.

Main Results:

• Simultaneous acquisition of high-resolution spatial, spectral, and fluorescence lifetime data.
• Demonstrated high fidelity in spatial, spectral, and lifetime measurements compared to ground truth.
• Achieved accelerated data acquisition through compressive sensing.
• Successful imaging of volumetric fluorescent samples with enhanced data richness.

Conclusions:

• The developed 3D multispectral fluorescence lifetime microscope offers a significant advancement in imaging capabilities.
• The system enables fast, non-invasive, and comprehensive characterization of fluorescent samples.
• This technology provides spectral and temporal information without compromising spatial resolution or increasing acquisition time.