Single capture quantitative oblique back-illumination microscopy

Area of Science:

• Biomedical Optics
• Microscopy
• Computational Imaging

Background:

• Quantitative oblique back-illumination microscopy (qOBM) enables label-free, 3D phase imaging of thick biological samples.
• Traditional qOBM requires multiple captures, limiting imaging speed and system simplicity.

Purpose of the Study:

• To develop a single-capture qOBM (SCqOBM) method using deep learning for faster, simpler phase recovery.
• To validate the accuracy and performance of SCqOBM in various biological applications.

Main Methods:

• Implemented a deep learning model for phase reconstruction from a single oblique back-illumination image.
• Applied SCqOBM to diverse biological samples, including in-vivo imaging of blood flow.
• Evaluated imaging speed for single-slice and volumetric refractive index tomography.

Main Results:

• SCqOBM accurately reconstructs phase information, comparable to traditional four-capture qOBM.
• Demonstrated non-invasive, in-vivo imaging of blood flow in mouse brain and human arm.
• Achieved high-speed quantitative phase imaging at 2 kHz and volumetric tomography at 10 volumes/second.

Conclusions:

• SCqOBM significantly enhances imaging speed and simplifies hardware requirements for quantitative phase imaging.
• The technique is suitable for dynamic, real-time applications and opens new possibilities in biomedical research and diagnostics.
• SCqOBM facilitates high-resolution, label-free imaging for non-invasive hematological assessments and in-vivo tissue analysis.