Large depth-of-field fluorescence microscopy based on deep learning supported by Fresnel incoherent correlation holography

Super-Resolution Image Optimisation Based on Gradient Iterative Fast Diffraction-Free Spot Algorithm

Diffraction significantly deteriorates the quality of the laser image, causing severe degradation that undermines the theoretical performance parameters of the autofocus system. In this paper, we conduct a comprehensive analysis of the non-uniform features of the images. To enhance the imaging quality of each individual image, we propose a de-diffraction algorithm based on gradient iteration. This algorithm is capable of rapidly removing the interference spots resulting from diffraction and restoring the distorted laser spots. By doing so, it effectively eliminates the inevitable reduction in the autofocus resolution and focusing accuracy caused by diffraction. Furthermore, the proposed calculation model for the intra-localisation interval significantly improves the convergence of the iterative calculation process. Through experiments, it has been verified that, under the same conditions, the interlayer resolution between the reflective surfaces of the samples processed using this algorithm is increased to a quarter of the original value. This remarkable improvement in resolution, which far exceeds the microscope's inherent resolution, demonstrates that the algorithm successfully achieves super-resolution for the microscope.

Development and performance validation of an affordable and portable high-resolution darkfield polarization-sensitive multispectral imaging microscope for the assessment of radiation dermatitis and fibrosis

Radiation therapy (RT) is widely used for cancer treatment but is found with side effects of radiation dermatitis and fibrosis thereby calling for timely assessment. Nevertheless, current clinical assessment methods are found to be subjective, prone to bias, and accompanied by variability. There is, therefore, an unmet clinical need to explore a new assessment technique, ideally portable and affordable, making it accessible to less developed regions too. We developed an affordable (16764 CNY) and portable high-resolution ((3.91 μm) darkfield polarization-sensitive multispectral imaging (PS-MSI) microscope. The implementation of the Monte Carlo simulation on the PS multi spectra allows the quantitative analysis of physiological parameters (i.e., blood volume fraction (BVF) and oxygen saturation of hemoglobin) at different skin layers for the dermatitis assessment. Further derivation of the degree of linear polarization (DOLP) reflects randomly distributed collagen fibers associated with fibrosis for the fibrosis assessment. PS-MSI microscope developed revealed a significant decrease (p < 0.001, analysis of variance, ANOVA) in the DOLP associated with fibrosis like scar tissue, and significant (p < 0.001, ANOVA) increases in BVF and oxygen saturation of hemoglobin accompanying artificially induced dermatitis. One-dimensional convolutional neural network implemented on the DOLP and multiple spectra achieved accuracies of 96% and 92.2%, respectively, for the classification of the artificially induced skin dermatitis and fibrosis like scar, demonstrating the potential of the affordable PS-MSI microscope developed for objective, unbiased and consistent assessment of radiation dermatitis and fibrosis in the clinics.

Research on the Method of Depth-Sensing Optical System Based on Multi-Layer Interface Reflection

In this paper, a depth-sensing method employing active irradiation of a semi-annular beam is proposed for observing the multi-layered reflective surfaces of transparent samples with higher resolutions and lower interference. To obtain the focusing resolution of the semi-annular aperture diaphragm system, a model for computing the diffracted optical energy distribution of an asymmetric aperture diaphragm is constructed, and mathematical formulas are deduced for determining the system resolution based on the position of the first dark ring of the amplitude distribution. Optical simulations were performed under specific conditions; the lateral resolution δr of the depth-sensing system was determined to be 0.68 μm, and the focusing accuracy δz was determined to be 0.60 μm. An experimental platform was established under the same conditions, and the results were in accord with those of the simulation results, which validated the correctness of the formula for calculating the amplitude distribution of the diffracted light from the asymmetric aperture diaphragm.

Fresnel incoherent compressive holography toward 3D videography via dual-channel simultaneous phase-shifting interferometry

Fresnel incoherent correlation holography (FINCH) enables high-resolution 3D imaging of objects from several 2D holograms under incoherent light and has many attractive applications in motionless 3D fluorescence imaging. However, FINCH has difficulty implementing 3D imaging of dynamic scenes since multiple phase-shifting holograms need to be recorded for removing the bias term and twin image in the reconstructed scene, which requires the object to remain static during this progress. Here, we propose a dual-channel Fresnel noncoherent compressive holography method. First, a pair of holograms with π phase shifts obtained in a single shot are used for removing the bias term noise. Then, a physic-driven compressive sensing (CS) algorithm is used to achieve twin-image-free reconstruction. In addition, we analyze the reconstruction effect and suitability of the CS algorithm and two-step phase-shift filtering algorithm for objects with different complexities. The experimental results show that the proposed method can record hologram videos of 3D dynamic objects and scenes without sacrificing the imaging field of view or resolution. Moreover, the system refocuses images at arbitrary depth positions via computation, hence providing a new method for fast high-throughput incoherent 3D imaging.

General phase-difference imaging of incoherent digital holography

The hologram formed by incoherent holography based on self-interference should preserve the phase difference information of the object, such as the phase difference between the mutually orthogonal polarizations of anisotropic object. How to decode this phase difference from this incoherent hologram, i.e., phase-difference imaging, is of great significance for studying the properties of the measured object. However, there is no general phase-difference imaging theory due to both diverse incoherent holography systems and the complicated reconstruction process from holograms based on the diffraction theory. To realize phase-difference image in incoherent holography, the relationship between the phase difference of the object and the image reconstructed by holograms is derived using a general physical model of incoherent holographic systems, and then the additional phase that will distort this relationship in actual holographic systems is analyzed and eliminated. Finally, the phase-difference imaging that is suitable for the most incoherent holographic systems is realized and the general theory is experimentally verified. This technology can be applied to phase-difference imaging of anisotropic objects, and has potential applications in materials science, biomedicine, polarized optics and other fields.

Imaging and Simulation of Ruthenium Derivative Coating Microbeads at the Opaque Electrode with Electrogenerated Chemiluminescence

Electrogenerated chemiluminescence (ECL) imaging is gaining increasing attention in various fields because of its high sensitivity, low background, and good temporal and spatial resolution. However, ECL imaging of microsized objects at the opaque electrode via top-view configuration is challenged with the reactants' diffusion and light propagation. Here, we imaged and numerically simulated ruthenium derivative coating polystyrene microbeads (Ru1-PS@MB) at the glassy carbon electrode (GCE) via top-view configuration by ECL imaging. The ruthenium derivative (bis(2,2'-bipyridine)-4'-methyl-4-carboxybipyridine-ruthenium N-succinimidyl ester-bis (hexafluorophosphate), Ru1), a typical ECL reagent, was covalently linked onto the surface of aminated PS@MBs via the amide reaction. "Strong emission in edge and weak emission in center" phenomena for fluorescence (FL) and ECL emissions were obtained from Ru1-PS@MB on GCE. Z-Stack imaging of the microsized Ru1-PS@MB luminescence was performed on GCE in the presence of tri-n-propylamine (TPA). It is found that the clear luminescence range of Ru1-PS@MB perpendicular to the electrode surface in ECL image is slightly smaller than that in the FL image. The bigger was the diameter of the microbeads (from 5 to 18 μm), the larger was the ECL luminescence range of Ru1-PS@MB perpendicular to the electrode surface (from 5 to 7 μm). Our findings, which are also supported by numerical simulation, provide insights into the ECL imaging of microsized objects at the electrode surface, which will raise promising ECL applications in bioassays and cell imaging at the microscale level.

Programmable focused laser differential interferometer with a spatial light modulator as a dynamic diffractive optical element

A spatial light modulator (SLM) is incorporated into a focused laser differential interferometer (FLDI) to generate a nonlinear array of beams, and this setup is used to measure the power spectral density of a Mach-1.5, underexpanded jet of air. The results are compared with measurements from a 1-point FLDI to assess the feasibility of using SLMs in FLDI to serve as dynamic diffractive elements for generating beam arrays of any shape. The spectra comparison illustrates that spatial light modulated-FLDI (SLM-FLDI) detects similar spectral profiles to that of 1-point FLDI, especially dominant frequencies in the jet. SLM-FLDI could provide a useful expansion of FLDI capabilities.

High-resolution multi-z confocal microscopy with a diffractive optical element

There has been recent interest in the development of fluorescence microscopes that provide high-speed volumetric imaging for life-science applications. For example, multi-z confocal microscopy enables simultaneous optically-sectioned imaging at multiple depths over relatively large fields of view. However, to date, multi-z microscopy has been hampered by limited spatial resolution owing to its initial design. Here we present a variant of multi-z microscopy that recovers the full spatial resolution of a conventional confocal microscope while retaining the simplicity and ease of use of our initial design. By introducing a diffractive optical element in the illumination path of our microscope, we engineer the excitation beam into multiple tightly focused spots that are conjugated to axially distributed confocal pinholes. We discuss the performance of this multi-z microscope in terms of resolution and detectability and demonstrate its versatility by performing in-vivo imaging of beating cardiomyocytes in engineered heart tissues and neuronal activity in c. elegans and zebrafish brains.

Single-shot Fresnel incoherent correlation holography via deep learning based phase-shifting technology

Fresnel incoherent correlation holography (FINCH) realizes non-scanning three-dimension (3D) images using spatial incoherent illumination, but it requires phase-shifting technology to remove the disturbance of the DC term and twin term that appears in the reconstruction field, thus increasing the complexity of the experiment and limits the real-time performance of FINCH. Here, we propose a single-shot Fresnel incoherent correlation holography via deep learning based phase-shifting (FINCH/DLPS) method to realize rapid and high-precision image reconstruction using only a collected interferogram. A phase-shifting network is designed to implement the phase-shifting operation of FINCH. The trained network can conveniently predict two interferograms with the phase shift of 2/3 π and 4/3 π from one input interferogram. Using the conventional three-step phase-shifting algorithm, we can conveniently remove the DC term and twin term of the FINCH reconstruction and obtain high-precision reconstruction through the back propagation algorithm. The Mixed National Institute of Standards and Technology (MNIST) dataset is used to verify the feasibility of the proposed method through experiments. In the test with the MNIST dataset, the reconstruction results demonstrate that in addition to high-precision reconstruction, the proposed FINCH/DLPS method also can effectively retain the 3D information by calibrating the back propagation distance in the case of reducing the complexity of the experiment, further indicating the feasibility and superiority of the proposed FINCH/DLPS method.

Large depth-of-field computational imaging with multi-spectral and dual-aperture optics

Large DOF (depth-of-field) with high SNR (signal-noise-ratio) imaging is a crucial technique for applications from security monitoring to medical diagnostics. However, traditional optical design for large DOF requires a reduction in aperture size, and hence with a decrease in light throughput and SNR. In this paper, we report a computational imaging system integrating dual-aperture optics with a physics-informed dual-encoder neural network to realize prominent DOF extension. Boosted by human vision mechanism and optical imaging law, the dual-aperture imaging system is consisted of a small-aperture NIR camera to provide sharp edge and a large-aperture VIS camera to provide faithful color. To solve the imaging inverse problem in NIR-VIS fusion with different apertures, a specific network with parallel double encoders and the multi-scale fusion module is proposed to adaptively extract and learn the useful features, which contributes to preventing color deviation while preserving delicate scene textures. The proposed imaging framework is flexible and can be designed in different protos with varied optical elements for different applications. We provide theory for system design, demonstrate a prototype device, establish a real-scene dataset containing 3000 images, perform elaborate ablation studies and conduct peer comparative experiments. The experimental results demonstrate that our method effectively produces high-fidelity with larger DOF range than input raw images about 3 times. Without complex optical design and strict practical limitations, this novel, intelligent and integratable system is promising for variable vision applications such as smartphone photography, computational measurement, and medical imaging.

FCE-Net: a fast image contrast enhancement method based on deep learning for biomedical optical images

Optical imaging is an important tool for exploring and understanding structures of biological tissues. However, due to the heterogeneity of biological tissues, the intensity distribution of the signal is not uniform and contrast is normally degraded in the raw image. It is difficult to be used for subsequent image analysis and information extraction directly. Here, we propose a fast image contrast enhancement method based on deep learning called Fast Contrast Enhancement Network (FCE-Net). We divided network into dual-path to simultaneously obtain spatial information and large receptive field. And we introduced the spatial attention mechanism to enhance the inter-spatial relationship. We showed that the cell counting task of mouse brain images processed by FCE-Net was with average precision rate of 97.6% ± 1.6%, and average recall rate of 98.4% ± 1.4%. After processing with FCE-Net, the images from vascular extraction (DRIVE) dataset could be segmented with spatial attention U-Net (SA-UNet) to achieve state-of-the-art performance. By comparing FCE-Net with previous methods, we demonstrated that FCE-Net could obtain higher accuracy while maintaining the processing speed. The images with size of 1024 × 1024 pixels could be processed by FCE-Net with 37fps based on our workstation. Our method has great potential for further image analysis and information extraction from large-scale or dynamic biomedical optical images.