Volumetric imaging by fluorescence microscopy is often limited by anisotropic spatial resolution, in which the axial resolution is inferior to the lateral resolution. To address this problem, we present a deep-learning-enabled unsupervised super-resolution technique that enhances anisotropic images in volumetric fluorescence microscopy. In contrast to the existing deep learning approaches that require matched high-resolution target images, our method greatly reduces the effort to be put into practice as the training of a network requires only a single 3D image stack, without a priori knowledge of the image formation process, registration of training data, or separate acquisition of target data. This is achieved based on the optimal transport-driven cycle-consistent generative adversarial network that learns from an unpaired matching between high-resolution 2D images in the lateral image plane and low-resolution 2D images in other planes. Using fluorescence confocal microscopy and light-sheet microscopy, we demonstrate that the trained network not only enhances axial resolution but also restores suppressed visual details between the imaging planes and removes imaging artifacts.

Nicotinamide adenine dinucleotide (NAD) exists in an oxidized form (NAD⁺) and a reduced form (NADH). NAD⁺ plays crucial roles in cancer metabolism, including in cellular signaling, energy production and redox regulation. However, it remains unclear whether NAD(H) pool size (NAD⁺ and NADH) could be used as biomarker for colon cancer progression. Here, we showed that the NAD(H) pool size and NAD⁺/NADH ratio both increased during colorectal cancer (CRC) progression due to activation of the NAD⁺ salvage pathway mediated by nicotinamide phosphoribosyltransferase (NAMPT). The NAMPT expression was upregulated in adenoma and adenocarcinoma tissues from CRC patients. The NADH fluorescence intensity measured by two‐photon excitation fluorescence (TPEF) microscopy was consistently increased in CRC cell lines, azoxymethane/dextran sodium sulfate (AOM/DSS)‐induced CRC tissues and tumor tissues from CRC patients. The increases in the NAD(H) pool inhibited the accumulation of excessive reactive oxygen species (ROS) levels and FK866, a specific inhibitor of NAMPT, treatment decreased the CRC nodule size by increasing ROS levels in AOM/DSS mice. Collectively, our results suggest that NAMPT‐mediated upregulation of the NAD(H) pool protects cancer cells against detrimental oxidative stress and that detecting NADH fluorescence by TPEF microscopy could be a potential method for monitoring CRC progression.

Myelination processes are closely related to higher brain functions such as learning and memory. While their longitudinal observation has been crucial to understanding myelin-related physiology and various brain disorders, skull opening or thinning has been required to secure clear optical access. Here we present a high-speed reflection matrix microscope using a light source with a wavelength of 1.3 μm to reduce tissue scattering and aberration. Furthermore, we develop a computational conjugate adaptive optics algorithm designed for the recorded reflection matrix to optimally compensate for the skull aberrations. These developments allow us to realize label-free longitudinal imaging of cortical myelin through an intact mouse skull. The myelination processes of the same mice were observed from 3 to 10 postnatal weeks to the depth of cortical layer 4 with a spatial resolution of 0.79 μm. Our system will expedite the investigations on the role of myelination in learning, memory, and brain disorders.