Deep self-learning enables fast, high-fidelity isotropic resolution restoration for volumetric fluorescence microscopy.

One intrinsic yet critical issue that troubles the field of fluorescence microscopy ever since its introduction is the unmatched resolution in the lateral and axial directions (i.e., resolution anisotropy), which severely deteriorates the quality, reconstruction, and analysis of 3D volume images. By leveraging the natural anisotropy, we present a deep self-learning method termed Self-Net that significantly improves the resolution of axial images by using the lateral images from the same raw dataset as rational targets. By incorporating unsupervised learning for realistic anisotropic degradation and supervised learning for high-fidelity isotropic recovery, our method can effectively suppress the hallucination with substantially enhanced image quality compared to previously reported methods. In the experiments, we show that Self-Net can reconstruct high-fidelity isotropic 3D images from organelle to tissue levels via raw images from various microscopy platforms, e.g., wide-field, laser-scanning, or super-resolution microscopy. For the first time, Self-Net enables isotropic whole-brain imaging at a voxel resolution of 0.2 × 0.2 × 0.2 µm3, which addresses the last-mile problem of data quality in single-neuron morphology visualization and reconstruction with minimal effort and cost. Overall, Self-Net is a promising approach to overcoming the inherent resolution anisotropy for all classes of 3D fluorescence microscopy.

Axial resolution and imaging contrast enhancement in inverted light-sheet microscopy by natural illumination modulation.

Inverted light-sheet microscopy (ILSM) is widely employed for fast large-volume imaging of biological tissue. However, the scattering especially in an uncleared sample, and the divergent propagation of the illumination beam lead to a trade-off between axial resolution and imaging depth. Herein, we propose naturally modulated ILSM (NM-ILSM) as a technique to improve axial resolution while simultaneously maintaining the wide field-of-view (FOV), and enhancing imaging contrast via background suppression. Theoretical derivations, simulations, and experimental imaging demonstrate 15% axial resolution increases, and fivefold greater image contrast compared with conventional ILSM. Therefore, NM-ILSM allows convenient imaging quality improvement for uncleared tissue and could extend the biological application scope of ILSM.

Multicolor high-resolution whole-brain imaging for acquiring and comparing the brain-wide distributions of type-specific and projection-specific neurons with anatomical annotation in the same brain.

Visualizing the relationships and interactions among different biological components in the whole brain is crucial to our understanding of brain structures and functions. However, an automatic multicolor whole-brain imaging technique is still lacking. Here, we developed a multicolor wide-field large-volume tomography (multicolor WVT) to simultaneously acquire fluorescent signals in blue, green, and red channels in the whole brain. To facilitate the segmentation of brain regions and anatomical annotation, we used 4', 6-diamidino-2-phenylindole (DAPI) to provide cytoarchitecture through real-time counterstaining. We optimized the imaging planes and modes of three channels to overcome the axial chromatic aberration of the illumination path and avoid the crosstalk from DAPI to the green channel without the modification of system configuration. We also developed an automatic contour recognition algorithm based on DAPI-staining cytoarchitecture to shorten data acquisition time and reduce data redundancy. To demonstrate the potential of our system in deciphering the relationship of the multiple components of neural circuits, we acquired and quantified the brain-wide distributions of cholinergic neurons and input of ventral Caudoputamen (CP) with the anatomical annotation in the same brain. We further identified the cholinergic type of upstream neurons projecting to CP through the triple-color collocated analysis and quantified its proportions in the two brain-wide distributions. Both accounted for 0.22%, implying CP might be modulated by non-cholinergic neurons. Our method provides a new research tool for studying the different biological components in the same organ and potentially facilitates the understanding of the processing mechanism of neural circuits and other biological activities.

Cryo-fluorescence micro-optical sectioning tomography for volumetric imaging of various whole organs with subcellular resolution.

Optical visualization of complex microstructures in the entire organ is essential for biomedical research. However, the existing methods fail to accurately acquire the detailed microstructures of whole organs with good morphological and biochemical preservation. This study proposes a cryo-fluorescence micro-optical sectioning tomography (cryo-fMOST) to image whole organs in three dimensions (3D) with submicron resolution. The system comprises a line-illumination microscope module, cryo-microtome, three-stage refrigeration module, and heat insulation device. To demonstrate the imaging capacity and wide applicability of the system, we imaged and reconstructed various organs of mice in 3D, including the healthy tongue, kidney, and brain, as well as the infarcted heart. More importantly, imaged brain slices were performed sugar phosphates determination and fluorescence in situ hybridization imaging to verify the compatibility of multi-omics measurements. The results demonstrated that cryo-fMOST is capable of acquiring high-resolution morphological details of various whole organs and may be potentially useful for spatial multi-omics.

EMDS-5: Environmental Microorganism image dataset Fifth Version for multiple image analysis tasks.

Environmental Microorganism Data Set Fifth Version (EMDS-5) is a microscopic image dataset including original Environmental Microorganism (EM) images and two sets of Ground Truth (GT) images. The GT image sets include a single-object GT image set and a multi-object GT image set. EMDS-5 has 21 types of EMs, each of which contains 20 original EM images, 20 single-object GT images and 20 multi-object GT images. EMDS-5 can realize to evaluate image preprocessing, image segmentation, feature extraction, image classification and image retrieval functions. In order to prove the effectiveness of EMDS-5, for each function, we select the most representative algorithms and price indicators for testing and evaluation. The image preprocessing functions contain two parts: image denoising and image edge detection. Image denoising uses nine kinds of filters to denoise 13 kinds of noises, respectively. In the aspect of edge detection, six edge detection operators are used to detect the edges of the images, and two evaluation indicators, peak-signal to noise ratio and mean structural similarity, are used for evaluation. Image segmentation includes single-object image segmentation and multi-object image segmentation. Six methods are used for single-object image segmentation, while k-means and U-net are used for multi-object segmentation. We extract nine features from the images in EMDS-5 and use the Support Vector Machine (SVM) classifier for testing. In terms of image classification, we select the VGG16 feature to test SVM, k-Nearest Neighbors, Random Forests. We test two types of retrieval approaches: texture feature retrieval and deep learning feature retrieval. We select the last layer of features of VGG16 network and ResNet50 network as feature vectors. We use mean average precision as the evaluation index for retrieval. EMDS-5 is available at the URL:https://github.com/NEUZihan/EMDS-5.git.