Review maps optical diffraction tomography for 3D refractive index imaging

A new review in Light: Advanced Manufacturing outlines how optical tomography and optical diffraction tomography reconstruct 3D refractive index maps from multi-angle wavefront data. The paper highlights uses in cell biology, pathology, materials science, and industrial inspection as the field moves toward faster, more accurate, AI-assisted imaging.

Why it matters: - Three-dimensional refractive index imaging gives quantitative views of transparent or weakly scattering samples without fluorescence labeling. - The method can capture morphology, density, stress, composition, and other internal properties in a single label-free workflow. - The approach has clear relevance for cell biology, pathology, hematology, microbiology, materials science, and industrial inspection.

What happened: - A review article in Light: Advanced Manufacturing summarizes optical tomography (OT) and optical diffraction tomography (ODT) for 3D refractive index imaging. - The review was written by the team of Professor Peng Gao and the team of Professor Chao Zuo. - The article covers principles, experimental implementations, reconstruction methods, representative applications, current trends, and future perspectives. - The paper is indexed under DOI 10.37188/lam.2026.077.

The details: - OT and ODT reconstruct 3D refractive index distributions from transmittance wavefronts recorded at multiple illumination angles. - The reconstruction can use Fourier slicing or diffraction principles. - OT uses the Fourier slicing theorem and treats the measured phase as the integral of refractive index along the projection direction. - OT is computationally efficient, but it does not account for diffraction effects. - ODT uses the Fourier diffraction theorem and models light-field propagation more accurately in weakly scattering samples. - ODT is aimed at higher-precision 3D refractive index reconstruction. - Common acquisition strategies include illumination rotation, sample rotation, and hybrid modulation. - Illumination rotation is gentler for live cells, but it is vulnerable to the missing-cone problem. - Sample rotation improves frequency coverage, but it requires more from sample morphology and mechanical stability. - Hybrid modulation can expand frequency coverage further, but it adds system complexity and acquisition time. - Complex-field acquisition methods fall into interference-based, refraction-based, and diffraction-based approaches. - Digital holographic microscopy is a common interference-based method because it offers high phase sensitivity. - Refraction-based methods provide stable systems with simpler structures. - Diffraction-based methods recover phase from intensity images and can simplify the system while suppressing coherent noise. - Reconstruction methods include analytical reconstruction, optimization-based reconstruction, and deep learning-based reconstruction. - Analytical methods are fast and physically transparent, but they rely on the weak-scattering assumption. - Optimization-based methods can model more complex propagation and scattering, but they demand more computation. - Deep learning methods show promise for missing-cone artifact suppression, resolution enhancement, and rapid reconstruction. - The review notes that deep learning still needs better generalization and interpretability. - After reconstruction, users can extract dry mass, volume, sphericity, and refractive index heterogeneity.

Between the lines: - The review frames 3D refractive index imaging as more than a microscopy tool, positioning it as a quantitative measurement platform. - The biggest technical tradeoff is clear: faster and simpler methods often sacrifice accuracy, while more complete reconstructions usually cost more in time or system complexity. - Deep learning is emerging as a practical fix for longstanding reconstruction limits, but the field still needs stronger physical grounding. - The breadth of applications suggests the same imaging core could serve both biology and non-destructive inspection of transparent materials.

What's next: - ODT is expected to move toward faster acquisition, more isotropic 3D reconstruction, and tighter integration with deep learning. - Future development will likely combine hardware modulation, complex-field acquisition, reconstruction algorithms, and multimodal imaging. - The review says ODT is on track to become an important tool for label-free 3D microscopic imaging and precision optical inspection.

The bottom line: - ODT and OT are converging into a label-free 3D imaging platform that can quantify internal structure in cells and materials, with AI likely to play a bigger role in making the technology faster and more accurate.