Lensfree Spectral Light-field Fusion Microscopy for Contrast- and Resolution-enhanced Imaging of Biological Specimens

TL;DR

This paper introduces a lensfree spectral light-field fusion microscopy system that enhances contrast and resolution in biological imaging by combining multispectral light encodings through Bayesian fusion, resulting in significant improvements in SNR, resolution, and contrast.

Contribution

The paper presents a novel multispectral lensfree microscopy system with Bayesian fusion that significantly improves image quality and resolution for biological specimens.

Findings

Achieved 6-12 dB SNR improvement

Realized six-fold reduction in dispersion index

Increased numerical aperture by ~16%

Abstract

A lensfree spectral light-field fusion microscopy (LSLFM) system is presented for enabling contrast- and resolution-enhanced imaging of biological specimens. LSLFM consists of a pulsed multispectral lensfree microscope for capturing interferometric light-field encodings at various wavelengths, and Bayesian-based fusion to reconstruct a fused object light-field from the encodings. By fusing unique object detail information captured at different wavelengths, LSLFM can achieve improved resolution, contrast, and signal-to-noise ratio (SNR) over a single-channel lensfree microscopy system. A five-channel LSLFM system was developed and quantitatively evaluated to validate the design. Experimental results demonstrated that the LSLFM system provided SNR improvements of 6-12 dB, as well as a six-fold improvement in the dispersion index (DI), over that achieved using a single-channel, resolution-enhancing lensfree deconvolution microscopy system or its multi-wavelength counterpart. Furthermore, the LSLFM system achieved an increase in numerical aperture (NA) of ~16% over a single-channel resolution-enhancing lensfree deconvolution microscopy system at the highest-resolution wavelength used in the study. Samples of Staurastrum paradoxum, a waterborne algae, and human corneal epithelial cells were imaged using the system to illustrate its potential for enhanced imaging of biological specimens.