Enhancing resolution and contrast in fibre bundle-based fluorescence microscopy using generative adversarial network.

Fibre bundle (FB)-based endoscopes are indispensable in biology and medical science due to their minimally invasive nature. However, resolution and contrast for fluorescence imaging are limited due to characteristic features of the FBs, such as low numerical aperture (NA) and individual fibre core sizes. In this study, we improved the resolution and contrast of sample fluorescence images acquired using in-house fabricated high-NA FBs by utilising generative adversarial networks (GANs). In order to train our deep learning model, we built an FB-based multifocal structured illumination microscope (MSIM) based on a digital micromirror device (DMD) which improves the resolution and the contrast substantially compared to basic FB-based fluorescence microscopes. After network training, the GAN model, employing image-to-image translation techniques, effectively transformed wide-field images into high-resolution MSIM images without the need for any additional optical hardware. The results demonstrated that GAN-generated outputs significantly enhanced both contrast and resolution compared to the original wide-field images. These findings highlight the potential of GAN-based models trained using MSIM data to enhance resolution and contrast in wide-field imaging for fibre bundle-based fluorescence microscopy. Lay Description: Fibre bundle (FB) endoscopes are essential in biology and medicine but suffer from limited resolution and contrast for fluorescence imaging. Here we improved these limitations using high-NA FBs and generative adversarial networks (GANs). We trained a GAN model with data from an FB-based multifocal structured illumination microscope (MSIM) to enhance resolution and contrast without additional optical hardware. Results showed significant enhancement in contrast and resolution, showcasing the potential of GAN-based models for fibre bundle-based fluorescence microscopy.

An LED-Based structured illumination microscope using a digital micromirror device and GPU accelerated image reconstruction.

When combined with computational approaches, fluorescence imaging becomes one of the most powerful tools in biomedical research. It is possible to achieve resolution figures beyond the diffraction limit, and improve the performance and flexibility of high-resolution imaging systems with techniques such as structured illumination microscopy (SIM) reconstruction. In this study, the hardware and software implementation of an LED-based super-resolution imaging system using SIM employing GPU accelerated parallel image reconstruction is presented. The sample is illuminated with two-dimensional sinusoidal patterns with various orientations and lateral phase shifts generated using a digital micromirror device (DMD). SIM reconstruction is carried out in frequency space using parallel CUDA kernel functions. Furthermore, a general purpose toolbox for the parallel image reconstruction algorithm and an infrastructure that allows all users to perform parallel operations on images without developing any CUDA kernel code is presented. The developed image reconstruction algorithm was run separately on a CPU and a GPU. Two different SIM reconstruction algorithms have been developed for the CPU as mono-thread CPU algorithm and multi-thread OpenMP CPU algorithm. SIM reconstruction of 1024 × 1024 px images was achieved in 1.49 s using GPU computation, indicating an enhancement by ∼28 and ∼20 in computation time when compared with mono-thread CPU computation and multi-thread OpenMP CPU computation, respectively.

Understanding the Link between Inflammasome and Apoptosis through the Response of THP-1 Cells against Drugs Using Droplet-Based Microfluidics.

Droplet-based microfluidic devices are used to investigate monocytic THP-1 cells in response to drug administration. Consistent and reproducible droplets are created, each of which acts as a bioreactor to carry out single cell experiments with minimized contamination and live cell tracking under an inverted fluorescence microscope for more than 2 days. Here, the effects of three different drugs (temsirolimus, rifabutin, and BAY 11-7082) on THP-1 are examined and the results are analyzed in the context of the inflammasome and apoptosis relationship. The ASC adaptor gene tagged with GFP is monitored as the inflammasome reporter. Thus, a systematic way is presented for deciphering cell-to-cell heterogeneity, which is an important issue in cancer treatment. The drug temsirolimus, which has effects of disrupting the mTOR pathway and triggering apoptosis in tumor cells, causes THP-1 cells to express ASC and to be involved in apoptosis. Treatment with rifabutin, which inhibits proliferation and initiates apoptosis in cells, affects ASC expression by first increasing and then decreasing it. CASP-3, which has a role in apoptosis and is directly related to ASC, has an increasing level in inflammasome conditioning. Thus, the cell under the effect of rifabutin might be faced with programmed cell death faster. The drug BAY 11-7082, which is responsible for NFκB inhibition, shows similar results to temsirolimus with more than 60% of cells having high fluorescence intensity (ASC expression). The microfluidic platform presented here offers strong potential for studying newly developed small-molecule inhibitors for personalized/precision medicine.

An Easy-to-Fabricate Microfluidic Shallow Trench Induced Three-Dimensional Cell Culturing and Imaging (STICI3D) Platform.

Compared to the established monolayer approach of two-dimensional cell cultures, three-dimensional (3D) cultures more closely resemble in vivo models; that is, the cells interact and form clusters mimicking their organization in native tissue. Therefore, the cellular microenvironment of these 3D cultures proves to be more clinically relevant. In this study, we present a novel easy-to-fabricate microfluidic shallow trench induced 3D cell culturing and imaging (STICI3D) platform, suitable for rapid fabrication as well as mass manufacturing. Our design consists of a shallow trench, within which various hydrogels can be formed in situ via capillary action, between and fully in contact with two side channels that allow cell seeding and media replenishment, as well as forming concentration gradients of various molecules. Compared to a micropillar-based burst valve design, which requires sophisticated microfabrication facilities, our capillary-based STICI3D can be fabricated using molds prepared with simple adhesive tapes and razors alone. The simple design supports the easy applicability of mass-production methods such as hot embossing and injection molding as well. To optimize the STICI3D design, we investigated the effect of individual design parameters such as corner radii, trench height, and surface wettability under various inlet pressures on the confinement of a hydrogel solution within the shallow trench using Computational Fluid Dynamics simulations supported with experimental validation. We identified ideal design values that improved the robustness of hydrogel confinement and reduced the effect of end-user dependent factors such as hydrogel solution loading pressure. Finally, we demonstrated cultures of human mesenchymal stem cells and human umbilical cord endothelial cells in the STICI3D to show that it supports 3D cell cultures and enables precise control of cellular microenvironment and real-time microscopic imaging. The easy-to-fabricate and highly adaptable nature of the STICI3D platform makes it suitable for researchers interested in fabricating custom polydimethylsiloxane devices as well as those who are in need of ready-to-use plastic platforms. As such, STICI3Ds can be used in imaging cell-cell interactions, angiogenesis, semiquantitative analysis of drug response in cells, and measurement of transport through cell sheet barriers.

Size-Based Sorting of Emulsion Droplets in Microfluidic Channels Patterned with Laser-Ablated Guiding Tracks.

We demonstrate an autonomous, high-throughput mechanism for sorting of emulsion droplets with different sizes concurrently flowing in a microfluidic Hele-Shaw channel. The aqueous droplets of varying radii suspended in olive oil are separated into different streamlines across the channel upon interaction with a shallow (depth ∼ 700 nm) inclined guiding track ablated into the polydimethylsiloxane-coated surface of the channel with focused femtosecond laser pulses. Specifically, the observed differences in the droplet trajectories along the guiding track arise due to the different scaling of the confinement force attracting the droplets into the track, fluid drag, and wall friction, with the droplet radius. In addition, the distance traveled by the droplets along the track also depends on the track width, with wider tracks providing more stable droplet guiding for any given droplet size. We systematically study the influence of the droplet size and velocity on the trajectory of the droplets in the channel and analyze the sensitivity of size-based droplet sorting for varying flow conditions. The droplet guiding and sorting experiments are complemented by modeling of the droplet motion in the channel flow using computational fluid dynamics simulations and a previously developed model of droplet guiding. Finally, we demonstrate a complete separation of droplets produced by fusion of two independent droplet streams at the inlet of the Hele-Shaw channel from unfused daughter droplets. The presented droplet sorting technique can find applications in the development of analytical and preparative microfluidic protocols.