A novel deep learning-based 3D cell segmentation framework for future image-based disease detection

Cell segmentation plays a crucial role in understanding, diagnosing, and treating diseases. Despite the recent success of deep learning-based cell segmentation methods, it remains challenging to accurately segment densely packed cells in 3D cell membrane images. Existing approaches also require fine-tuning multiple manually selected hyperparameters on the new datasets. We develop a deep learning-based 3D cell segmentation pipeline, 3DCellSeg, to address these challenges. Compared to the existing methods, our approach carries the following novelties: (1) a robust two-stage pipeline, requiring only one hyperparameter; (2) a light-weight deep convolutional neural network (3DCellSegNet) to efficiently output voxel-wise masks; (3) a custom loss function (3DCellSeg Loss) to tackle the clumped cell problem; and (4) an efficient touching area-based clustering algorithm (TASCAN) to separate 3D cells from the foreground masks. Cell segmentation experiments conducted on four different cell datasets show that 3DCellSeg outperforms the baseline models on the ATAS (plant), HMS (animal), and LRP (plant) datasets with an overall accuracy of 95.6%, 76.4%, and 74.7%, respectively, while achieving an accuracy comparable to the baselines on the Ovules (plant) dataset with an overall accuracy of 82.2%. Ablation studies show that the individual improvements in accuracy is attributable to 3DCellSegNet, 3DCellSeg Loss, and TASCAN, with the 3DCellSeg demonstrating robustness across different datasets and cell shapes. Our results suggest that 3DCellSeg can serve a powerful biomedical and clinical tool, such as histo-pathological image analysis, for cancer diagnosis and grading.

ZELDA: A 3D Image Segmentation and Parent-Child Relation Plugin for Microscopy Image Analysis in napari

Bioimage analysis workflows allow the measurement of sample properties such as fluorescence intensity and polarization, cell number, and vesicles distribution, but often require the integration of multiple software tools. Furthermore, it is increasingly appreciated that to overcome the limitations of the 2D-view-based image analysis approaches and to correctly understand and interpret biological processes, a 3D segmentation of microscopy data sets becomes imperative. Despite the availability of numerous algorithms for the 2D and 3D segmentation, the latter still offers some challenges for the end-users, who often do not have either an extensive knowledge of the existing software or coding skills to link the output of multiple tools. While several commercial packages are available on the market, fewer are the open-source solutions able to execute a complete 3D analysis workflow. Here we present ZELDA, a new napari plugin that easily integrates the cutting-edge solutions offered by python ecosystem, such as scikit-image for image segmentation, matplotlib for data visualization, and napari multi-dimensional image viewer for 3D rendering. This plugin aims to provide interactive and zero-scripting customizable workflows for cell segmentation, vesicles counting, parent-child relation between objects, signal quantification, and results presentation; all included in the same open-source napari viewer, and “few clicks away”.

Live-cell microscopy or fluorescence anisotropy with budded baculoviruses - which way to go with measuring ligand binding to M4 muscarinic receptors?

M4 muscarinic receptor is a G protein-coupled receptor that has been associated with alcohol and cocaine abuse, Alzheimer's disease and schizophrenia which makes it an interesting drug target. For many G protein-coupled receptors, the development of high-affinity fluorescence ligands has expanded the options for high throughput screening of drug candidates and serve as useful tools in fundamental receptor research. So far, the lack of suitable fluorescence ligands has limited studying M4 receptor ligand binding. Here, we explored the possibilities of using fluorescence-based methods for studying binding affinity and kinetics to M4 receptor of both labeled and unlabeled ligands. We used two TAMRA-labeled fluorescence ligands, UR-MK342 and UR-CG072, for assay development. Using budded baculovirus particles as M4 receptor preparation and fluorescence anisotropy method, we determined the affinities and binding kinetics of both fluorescence ligands. The fluorescence ligands could also be used as reported probes for determining binding affinities of a set of unlabeled ligands. Based on these results, we took a step further towards a more natural signaling system and developed a method using live CHO-K1-hM4R cells and automated fluorescence microscopy suitable for routine determination of unlabeled ligand affinities. For quantitative image analysis, we developed random forest and deep learning-based pipelines for cell segmentation. The pipelines were integrated into the user-friendly open-source Aparecium software. All developed assays were suitable for measuring fluorescence ligand saturation binding, association and dissociation kinetics as well as for screening binding affinities of unlabeled ligands.

Using Imaging Mass Cytometry to Define Cell Identities and Interactions in Human Tissues

In the evolving landscape of highly multiplexed imaging techniques that can be applied to study complex cellular microenvironments, this review characterizes the use of imaging mass cytometry (IMC) to study the human kidney. We provide technical details for antibody validation, cell segmentation, and data analysis specifically tailored to human kidney samples, and elaborate on phenotyping of kidney cell types and novel insights that IMC can provide regarding pathophysiological processes in the injured or diseased kidney. This review will provide the reader with the necessary background to understand both the power and the limitations of IMC and thus support better perception of how IMC analysis can improve our understanding of human disease pathogenesis and can be integrated with other technologies such as single cell sequencing and proteomics to provide spatial context to cellular data.

Scellseg: a style-aware cell instance segmentation tool with pre-training and contrastive fine-tuning

Deep learning-based cell segmentation is increasingly utilized in cell biology and molecular pathology, due to massive accumulation of diverse large-scale datasets and excellent performance in cell representation. However, the development of specialized algorithms has long been hampered by a paucity of annotated training data, whereas the performance of generalist algorithm was limited without experiment-specific calibration. Here, we present a deep learning-based tool called Scellseg consisted of novel pre-trained network architecture and contrastive fine-tuning strategy. In comparison to four commonly used algorithms, Scellseg outperformed in average precision on three diverse datasets with no need for dataset-specific configuration. Interestingly, we found that eight images are sufficient for model tuning to achieve satisfied performance based on a shot data scale experiment. We also developed a graphical user interface integrated with functions of annotation, fine-tuning and inference, that allows biologists to easily specialize their own segmentation model and analyze data at the single-cell level.

Fully Automatic Cell Segmentation with Fourier Descriptors

Cell segmentation is a fundamental problem in biology for which convolutional neural networks yield the best results nowadays. In this paper, we present HarmonicNet, a network, which is a modification of the popular StarDist and SplineDist architectures. While StarDist and SplineDist describe an object by the lengths of equiangular rays and control points respectively, our network utilizes Fourier descriptors, predicting a coefficient vector for every pixel on the image, which implicitly define the resulting segmentation. We evaluate our model on three different datasets, and show that Fourier descriptors can achieve a high level of accuracy with a small number of coefficients. HarmonicNet is also capable of accurately segmenting objects that are not star-shaped, a case where StarDist performs suboptimally according to our experiments.

Training a deep learning model for single-cell segmentation without manual annotation

Advances in the artificial neural network have made machine learning techniques increasingly more important in image analysis tasks. Recently, convolutional neural networks (CNN) have been applied to the problem of cell segmentation from microscopy images. However, previous methods used a supervised training paradigm in order to create an accurate segmentation model. This strategy requires a large amount of manually labeled cellular images, in which accurate segmentations at pixel level were produced by human operators. Generating training data is expensive and a major hindrance in the wider adoption of machine learning based methods for cell segmentation. Here we present an alternative strategy that trains CNNs without any human-labeled data. We show that our method is able to produce accurate segmentation models, and is applicable to both fluorescence and bright-field images, and requires little to no prior knowledge of the signal characteristics.