A Weakly Supervised Learning Method for Cell Detection and Tracking Using Incomplete Initial Annotations

The automatic detection of cells in microscopy image sequences is a significant task in biomedical research. However, routine microscopy images with cells, which are taken during the process whereby constant division and differentiation occur, are notoriously difficult to detect due to changes in their appearance and number. Recently, convolutional neural network (CNN)-based methods have made significant progress in cell detection and tracking. However, these approaches require many manually annotated data for fully supervised training, which is time-consuming and often requires professional researchers. To alleviate such tiresome and labor-intensive costs, we propose a novel weakly supervised learning cell detection and tracking framework that trains the deep neural network using incomplete initial labels. Our approach uses incomplete cell markers obtained from fluorescent images for initial training on the Induced Pluripotent Stem (iPS) cell dataset, which is rarely studied for cell detection and tracking. During training, the incomplete initial labels were updated iteratively by combining detection and tracking results to obtain a model with better robustness. Our method was evaluated using two fields of the iPS cell dataset, along with the cell detection accuracy (DET) evaluation metric from the Cell Tracking Challenge (CTC) initiative, and it achieved 0.862 and 0.924 DET, respectively. The transferability of the developed model was tested using the public dataset FluoN2DH-GOWT1, which was taken from CTC; this contains two datasets with reference annotations. We randomly removed parts of the annotations in each labeled data to simulate the initial annotations on the public dataset. After training the model on the two datasets, with labels that comprise 10% cell markers, the DET improved from 0.130 to 0.903 and 0.116 to 0.877. When trained with labels that comprise 60% cell markers, the performance was better than the model trained using the supervised learning method. This outcome indicates that the model's performance improved as the quality of the labels used for training increased.

Tracking Permeation of Dimethyl Sulfoxide (DMSO) in Mentha × piperita Shoot Tips Using Coherent Raman Microscopy

Cryopreservation has emerged as a low-maintenance, cost-effective solution for the long-term preservation of vegetatively propagated crops. Shoot tip cryopreservation often makes use of vitrification methods that employ highly concentrated mixtures of cryoprotecting agents; however, little is understood as to how these cryoprotecting agents protect cells and tissues from freezing. In this study, we use coherent anti-Stokes Raman scattering microscopy to directly visualize where dimethyl sulfoxide (DMSO) localizes within Mentha × piperita shoot tips. We find that DMSO fully penetrates the shoot tip tissue within 10 min of exposure. Variations in signal intensities across images suggest that DMSO may interact with cellular components, leading to its accumulation in specific regions.

Label-Free Imaging Analysis of Patient-Derived Cholangiocarcinoma Organoids after Sorafenib Treatment

Monitoring tumor growth dynamics is crucial for understanding cancer. To establish an in vitro method for the continuous assessment of patient-specific tumor growth, tumor organoids were generated from patients with intrahepatic CCA (iCCA). Organoid growth was monitored for 48 h by label-free live brightfield imaging. Growth kinetics were calculated and validated by MTS assay as well as immunohistochemistry of Ki67 to determine proliferation rates. We exposed iCCA organoids (iCCAOs) and non-tumor intrahepatic cholangiocyte organoids (ICOs) to sub-therapeutic concentrations of sorafenib. Monitoring the expansion rate of iCCAOs and ICOs revealed that iCCAO growth was inhibited by sorafenib in a time- and dose-dependent fashion, while ICOs were unaffected. Quantification of the proliferation marker Ki67 confirmed inhibition of iCCAO growth by roughly 50% after 48 h of treatment with 4 µM sorafenib. We established a robust analysis pipeline combining brightfield microscopy and a straightforward image processing approach for the label-free growth monitoring of patient-derived iCCAOs. Combined with bioanalytical validation, this approach is suitable for a fast and efficient high-throughput drug screening in tumor organoids to develop patient-specific systemic treatment options.

Auer rods and faggot cells: A review of the history, significance and mimics of two morphological curiosities of enduring relevance

Myeloid differentiation in blasts is distinguished by the presence of one or more needle-shaped crystals called Auer rods. Auer rods manifest either alone or as faggot cells (containing bundles of Auer rods) in various types of acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and myelodysplastic/myeloproliferative neoplasms (MDS/MPN). Their presence largely portends a better prognosis in AML (as markers of maturation/differentiation), and upstages cases of MDS and MDS/MPN. Observation of these rods in residual blasts in treated cases of AML indicate an absence of remission. This article traces their historical discovery and elucidation of their pathogenetic intricacies, as well as our current understanding of their relevance in myeloid neoplasms. Studies evaluating their prognostic impact in AML and MDS are catalogued. We also discuss a variety of other hematological and non-hematological neoplasms where structures potentially mistakeable with Auer rods have been described. Even as the diagnostic approach to hematological malignancies has evolved from a morphology+cytochemistry+immunophenotyping-dependent one in the last century to a predominantly molecular genetics-based classification currently, and even as high-throughput sequencing and structural variation detection techniques surpass morphology in detecting clinically-relevant sub-categories of similar-appearing tumours, we review these curious microscopic structures that have withstood the test of time with respect to their diagnostic relevance. This article is protected by copyright. All rights reserved.

Evaluating Very Deep Convolutional Neural Networks for Nucleus Segmentation from Brightfield Cell Microscopy Images

Advances in microscopy have increased output data volumes, and powerful image analysis methods are required to match. In particular, finding and characterizing nuclei from microscopy images, a core cytometry task, remains difficult to automate. While deep learning models have given encouraging results on this problem, the most powerful approaches have not yet been tested for attacking it. Here, we review and evaluate state-of-the-art very deep convolutional neural network architectures and training strategies for segmenting nuclei from brightfield cell images. We tested U-Net as a baseline model; considered U-Net++, Tiramisu, and DeepLabv3+ as latest instances of advanced families of segmentation models; and propose PPU-Net, a novel light-weight alternative. The deeper architectures outperformed standard U-Net and results from previous studies on the challenging brightfield images, with balanced pixel-wise accuracies of up to 86%. PPU-Net achieved this performance with 20-fold fewer parameters than the comparably accurate methods. All models perform better on larger nuclei and in sparser images. We further confirmed that in the absence of plentiful training data, augmentation and pretraining on other data improve performance. In particular, using only 16 images with data augmentation is enough to achieve a pixel-wise F1 score that is within 5% of the one achieved with a full data set for all models. The remaining segmentation errors are mainly due to missed nuclei in dense regions, overlapping cells, and imaging artifacts, indicating the major outstanding challenges.

Simple Shading Correction Method for Brightfield Whole Slide Imaging

Whole slide imaging (WSI) refers to the process of creating a high-resolution digital image of a whole slide. Since digital images are typically produced by stitching image sequences acquired from different fields of view, the visual quality of the images can be degraded owing to shading distortion, which produces black plaid patterns on the images. A shading correction method for brightfield WSI is presented, which is simple but robust not only against typical image artifacts caused by specks of dust and bubbles, but also against fixed-pattern noise, or spatial variations in pixel values under uniform illumination. The proposed method comprises primarily of two steps. The first step constructs candidates of a shading distortion model from a stack of input image sequences. The second step selects the optimal model from the candidates. The proposed method was compared experimentally with two previous state-of-the-art methods, regularized energy minimization (CIDRE) and background and shading correction (BaSiC) and showed better correction scores, as smooth operations and constraints were not imposed when estimating the shading distortion. The correction scores, averaged over 40 image collections, were as follows: proposed method, 0.39 ± 0.099; CIDRE method, 0.67 ± 0.047; BaSiC method, 0.55 ± 0.038. Based on the quantitative evaluations, we can confirm that the proposed method can correct not only shading distortion, but also fixed-pattern noise, compared with the two previous state-of-the-art methods.

Ultrastructural evaluation of mesenchymal stem cells from inflamed periodontium in different in vitro conditions

This research aimed to observe the behavior of mesenchymal stem cells (MSCs) isolated from periodontal granulation tissue (gt) when manipulated ex vivo to induce three-dimensional (3D) spheroid (aggregates) formation as well as when seeded on two bone scaffolds of animal origin. Periodontal gt was chosen as a MSC source because of its availability, considering that it is eliminated as a waste material during conventional surgical therapies. 3D aggregates of cells were generated; they were grown for 3 and 7 days, respectively, and then prepared for transmission electron microscopic analysis. The two biomaterials were seeded for 72 h with gtMSCs and prepared for scanning electronic microscopic observation. The ultrastructural analysis of 3D spheroids remarked some differences between the inner and the outer cell layers, with a certain commitment observed at the inner cells. Both scaffolds showed a relatively smooth surface at low magnification. Macro- and micropores having a scarce distribution were observed on both bone substitutes. gtMSCs grew with relative difficulty on the biomaterials. After 72 h of proliferation, gtMSCs scarcely covered the surface of bovine bone scaffolds, demonstrating fibroblast-like or star-like shapes with elongated filiform extensions. Our results add other data on the possible usefulness of gtMSC and could question the current paradigm regarding the complete removal of chronically inflamed gts from the defects during periodontal surgeries. Until optimal protocols for ex vivo manipulation of MSCs are available for clinical settings, it is advisable to use biocompatible bone substitutes that allow the development of progenitor cells.

3D cryo-imaging: a very high-resolution view of the whole mouse

We developed the Case Cryo-imaging system that provides information rich, very high-resolution, color brightfield, and molecular fluorescence images of a whole mouse using a section-and-image block-face imaging technology. The system consists of a mouse-sized, motorized cryo-microtome with special features for imaging, a modified, brightfield/fluorescence microscope, and a robotic xyz imaging system positioner, all of which is fully automated by a control system. Using the robotic system, we acquired microscopic tiled images at a pixel size of 15.6 microm over the block face of a whole mouse sectioned at 40 microm, with a total data volume of 55 GB. Viewing 2D images at multiple resolutions, we identified small structures such as cardiac vessels, muscle layers, villi of the small intestine, the optic nerve, and layers of the eye. Cryo-imaging was also suitable for imaging embryo mutants in 3D. A mouse, in which enhanced green fluorescent protein was expressed under gamma actin promoter in smooth muscle cells, gave clear 3D views of smooth muscle in the urogenital and gastrointestinal tracts. With cryo-imaging, we could obtain 3D vasculature down to 10 microm, over very large regions of mouse brain. Software is fully automated with fully programmable imaging/sectioning protocols, email notifications, and automatic volume visualization. With a unique combination of field-of-view, depth of field, contrast, and resolution, the Case Cryo-imaging system fills the gap between whole animal in vivo imaging and histology.

Whole Mouse Cryo-Imaging

The Case cryo-imaging system is a section and image system which allows one to acquire micron-scale, information rich, whole mouse color bright field and molecular fluorescence images of an entire mouse. Cryo-imaging is used in a variety of applications, including mouse and embryo anatomical phenotyping, drug delivery, imaging agents, metastastic cancer, stem cells, and very high resolution vascular imaging, among many. Cryo-imaging fills the gap between whole animal in vivo imaging and histology, allowing one to image a mouse along the continuum from the mouse → organ → tissue structure → cell → sub-cellular domains. In this overview, we describe the technology and a variety of exciting applications. Enhancements to the system now enable tiled acquisition of high resolution images to cover an entire mouse. High resolution fluorescence imaging, aided by a novel subtraction processing algorithm to remove sub-surface fluorescence, makes it possible to detect fluorescently-labeled single cells. Multi-modality experiments in Magnetic Resonance Imaging and Cryo-imaging of a whole mouse demonstrate superior resolution of cryo-images and efficiency of registration techniques. The 3D results demonstrate the novel true-color volume visualization tools we have developed and the inherent advantage of cryo-imaging in providing unlimited depth of field and spatial resolution. The recent results continue to demonstrate the value cryo-imaging provides in the field of small animal imaging research.