Search for: All records

Imaging through scattering is a pervasive and difficult problem in many biological applications. The high background and the exponentially attenuated target signals due to scattering fundamentally limits the imaging depth of fluorescence microscopy. Light-field systems are favorable for high-speed volumetric imaging, but the 2D-to-3D reconstruction is fundamentally ill-posed, and scattering exacerbates the condition of the inverse problem. Here, we develop a scattering simulator that models low-contrast target signals buried in heterogeneous strong background. We then train a deep neural network solely on synthetic data to descatter and reconstruct a 3D volume from a single-shot light-field measurement with low signal-to-background ratio (SBR). We apply this network to our previously developed computational miniature mesoscope and demonstrate the robustness of our deep learning algorithm on scattering phantoms with different scattering conditions. The network can robustly reconstruct emitters in 3D with a 2D measurement of SBR as low as 1.05 and as deep as a scattering length. We analyze fundamental tradeoffs based on network design factors and out-of-distribution data that affect the deep learning model’s generalizability to real experimental data. Broadly, we believe that our simulator-based deep learning approach can be applied to a wide range of imaging through scattering techniques where experimental paired training data is lacking.

 

We propose a novel algorithm based on the split-step non-paraxial model for different intensity diffraction tomography setups to recover the 3D refractive index distribution of multiple-scattering biological samples.

 

We report a new technique for single-shot quantitative phase retrieval from transparent objects, based on plasmonic metasurface photodetectors featuring an asymmetric angular response around normal incidence combined with a particularly simple optical setup. 

Abstract The visualization of pure phase objects by wavefront sensing has important applications ranging from surface profiling to biomedical microscopy, and generally requires bulky and complicated setups involving optical spatial filtering, interferometry, or structured illumination. Here we introduce a new type of image sensors that are uniquely sensitive to the local direction of light propagation, based on standard photodetectors coated with a specially designed plasmonic metasurface that creates an asymmetric dependence of responsivity on angle of incidence around the surface normal. The metasurface design, fabrication, and angle-sensitive operation are demonstrated using a simple photoconductive detector platform. The measurement results, combined with computational imaging calculations, are then used to show that a standard camera or microscope based on these metasurface pixels can directly visualize phase objects without any additional optical elements, with state-of-the-art minimum detectable phase contrasts below 10 mrad. Furthermore, the combination of sensors with equal and opposite angular response on the same pixel array can be used to perform quantitative phase imaging in a single shot, with a customized reconstruction algorithm which is also developed in this work. By virtue of its system miniaturization and measurement simplicity, the phase imaging approach enabled by these devices is particularly significant for applications involving space-constrained and portable setups (such as point-of-care imaging and endoscopy) and measurements involving freely moving objects. 

We report the development of angle-sensitive photodetectors based on specially designed metasurfaces that can map the phase distribution of the incident light and visualize transparent phase objects without any external spatial-filtering elements. Pixel arrays of these devices can provide quantitative phase reconstruction in a single shot with state-of-the-art sensitivity. 

We report plasmonic metasurface photodetectors featuring a strong asymmetric angular response around normal incidence that can visualize transparent phase objects with high sensitivity in a simple and compact imaging setup.

 

Recovering 3D phase features of complex biological samples traditionally sacrifices computational efficiency and processing time for physical model accuracy and reconstruction quality. Here, we overcome this challenge using an approximant-guided deep learning framework in a high-speed intensity diffraction tomography system. Applying a physics model simulator-based learning strategy trained entirely on natural image datasets, we show our network can robustly reconstruct complex 3D biological samples. To achieve highly efficient training and prediction, we implement a lightweight 2D network structure that utilizes a multi-channel input for encoding the axial information. We demonstrate this framework on experimental measurements of weakly scattering epithelial buccal cells and strongly scatteringC. elegansworms. We benchmark the network’s performance against a state-of-the-art multiple-scattering model-based iterative reconstruction algorithm. We highlight the network’s robustness by reconstructing dynamic samples from a living worm video. We further emphasize the network’s generalization capabilities by recovering algae samples imaged from different experimental setups. To assess the prediction quality, we develop a quantitative evaluation metric to show that our predictions are consistent with both multiple-scattering physics and experimental measurements.

 

We propose a novel intensity diffraction tomography (IDT) reconstruction algorithm based on the split-step non-paraxial (SSNP) model for recovering the 3D refractive index (RI) distribution of multiple-scattering biological samples. High-quality IDT reconstruction requires high-angle illumination to encode both low- and high- spatial frequency information of the 3D biological sample. We show that our SSNP model can more accurately compute multiple scattering from high-angle illumination compared to paraxial approximation-based multiple-scattering models. We apply this SSNP model to both sequential and multiplexed IDT techniques. We develop a unified reconstruction algorithm for both IDT modalities that is highly computationally efficient and is implemented by a modular automatic differentiation framework. We demonstrate the capability of our reconstruction algorithm on both weakly scattering buccal epithelial cells and strongly scattering liveC. elegansworms and liveC. elegansembryos.

 

We demonstrate an adaptive learning framework, termed dynamic synthesis network (DSN), which dynamically synthesizes model weights and adapts to different scattering conditions. The efficiency of DSN is demonstrated in holographic 3D particle descattering.