Cell Image Analysis Techniques

Depositing biological segmentation datasets FAIRly

This paper reviews existing biological segmentation datasets and recommends improvements to make them more reusable following FAIR principles.

In toto analysis of multicellular arrangement reduces embryonic tissue diversity to two archetypes that require specific cadherin expression

The paper introduces nuQLOUD, a framework to analyze tissue architecture, revealing two archetypes in zebrafish embryos driven by cadherin expression.

SubCell: Proteome-aware vision foundation models for microscopy capture single-cell biology

SubCell is a deep learning model for fluorescence microscopy that captures cellular morphology and protein organization, outperforming existing methods and enabling a multimodal approach to gene function.

deepBlastoid: A Deep Learning-Based High-Throughput Classifier for Human Blastoids Using Brightfield Images with Confidence Assessment

deepBlastoid is a deep learning model that classifies human blastoid brightfield images into quality categories with high accuracy and throughput, outperforming human experts in speed while matching their accuracy.

Video tracking of single cells to identify clustering behaviour

This paper introduces a graph theory-based single-cell tracking method to identify clustering behavior in PC3 cells, finding that co-cultured cells form persistent clusters while monocultured cells do not.

Systematic analysis of immune cell motility leveraging Immunemap, an open intravital microscopy atlas

The paper introduces Immunemap, an open intravital microscopy atlas, and uses unsupervised learning to identify four immune cell motility patterns, including previously uncharacterized behaviors.

Protocol for open-source Automated Universal high-content multiplex fluorescence for RNA in situ Analysis (AURA)

The paper describes an open-source protocol for automated high-content multiplex fluorescence in situ transcriptomics analysis using Fiji macros and a Python app for data summarization.

A tool to automate assessment of regional brain atrophy in mouse models of neurodegenerative disease

The paper introduces a tool to automate regional brain atrophy assessment in mouse models, showing high agreement with manual methods and reduced variability.

Classification of iPSC-Derived Cultures Using Convolutional Neural Networks to Identify Single Differentiated Neurons for Isolation or Measurement

The paper presents a CNN-based model for identifying single neurons in iPSC-derived cultures with high accuracy, and demonstrates its applicability to other cell types.

ReSCU-Nets: recurrent U-Nets for segmentation of multidimensional microscopy data

ReSCU-Nets improve segmentation of multidimensional microscopy data by using previous frame results as prompts, outperforming existing models in time-lapse sequences.

SpatialQPFs: An R package for deciphering cell-cell spatial relationship

SpatialQPFs provides an R package for extracting interpretable spatial features from cell imaging data using spatial statistical methods, enabling comprehensive spatial analyses across diverse contexts.

UniSPAC: A Unified Segmentation Framework for Proofreading and Annotation in Connectomics

UniSPAC improves neuronal segmentation in connectomics by combining 2D interactive proofreading and 3D tracing, reducing manual effort and enabling semi-automatic labeling.

Label-free multiplex microscopic imaging by image-to-image translation overcoming the trade-off between pixel- and image-level similarity

This paper develops a new image-to-image translation model for label-free multiplex microscopic imaging, achieving high performance at both pixel and image levels.

Global Neuron Shape Reasoning with Point Affinity Transformers

This paper introduces a point affinity transformer framework for global neuron shape reasoning, improving segmentation error correction and neuron type classification in connectomics.

SPACe (Swift Phenotypic Analysis of Cells): an open-source, single cell analysis of Cell Painting data

SPACe is an open-source tool for analyzing Cell Painting data, offering faster processing and better handling of cell-to-cell variation compared to existing software.

CelloType: A Unified Model for Segmentation and Classification of Tissue Images

CelloType is a multi-task learning model for cell segmentation and classification in tissue images, improving accuracy over existing methods.

Understanding TCR T cell knockout behavior using interpretable machine learning

This study uses explainable AI to analyze live-cell imaging data and identify behavioral changes in TCR T cells with specific genetic knockouts.

Cellular behavior analysis from live-cell imaging of TCR T cell-cancer cell interactions

The paper develops tools for analyzing T cell behavior from live-cell imaging to improve T cell therapies by studying the effects of genetic modifications on T cell-cancer cell interactions.

U-Net Ensembles for Segmentation of High-Density Cell Populations and Organelles

The paper develops a U-Net ensemble method for segmenting cells and organelles in 3D vEM data, achieving state-of-the-art results on CREMI neuron data and murine thrombus platelets.

Combining NeuroPainting with transcriptomics reveals cell-type-specific morphological and molecular signatures of the 22q11.2 deletion

The paper introduces NeuroPainting, a morphological phenotyping method, and combines it with transcriptomics to identify cell-type-specific effects of the 22q11.2 deletion on mitochondrial structure and gene expression in astrocytes.

ImmunoCellCycle-ID: A high-precision immunofluorescence-based method for cell cycle identification

This paper introduces an immunofluorescence-based method for high-precision cell cycle identification at the single-cell level, offering an alternative to flow cytometry for analyzing cell cycle stages.

Analysis Pipeline to Quantify Uterine Gland Structural Variations

The paper presents an image analysis pipeline to quantify structural variations in uterine glands across species, revealing shape changes during embryo implantation.

BIFROST: a method for registering diverse imagingdatasets of the Drosophila brain

BIFROST registers diverse Drosophila brain imaging datasets into a common anatomical space, enabling cross-experiment and cross-modality comparisons with sub-10 micron precision.

EasyFlow: An open source, user friendly cytometry analyzer with graphic user interface (GUI)

EasyFlow provides an open-source, user-friendly flow cytometry analysis tool with a GUI, accessible without coding knowledge and supporting both Matlab and Python.

Capturing cell heterogeneity in representations of cell populations for image-based profiling using contrastive learning

The paper introduces CytoSummaryNet, a contrastive learning method for aggregating single-cell features in image-based profiling, improving mechanism of action prediction by 30-68%.

The Brain Image Library: A Community-Contributed Microscopy Resource for Neuroscientists

The paper introduces the Brain Image Library (BIL), a centralized resource for brain microscopy data that enables exploration, visualization, and analysis without data download.

Deep Learning-driven Automatic Nuclei Segmentation of Label-free Live Cell Chromatin-sensitive Partial Wave Spectroscopic Microscopy Imaging

This paper introduces csPWS-seg, a deep learning model for automated nuclei segmentation in label-free live cell csPWS microscopy, achieving high accuracy with median IoU 0.80 and DSC 0.88.

Automated cell detection for immediate early gene-expressing neurons using inhomogeneous background subtraction in fluorescent images

This paper introduces ADABA, an automated cell detection method for fluorescent images with inhomogeneous backgrounds, showing it matches manual detection in mouse brain regions.

OrganoLabeling: Quick and accurate annotation tool for organoid images

OrganoLabeling automates image segmentation for organoid datasets, achieving segmentation accuracy comparable to manual labeling and improving efficiency.

LUMIC: Latent diffUsion for Multiplexed Images of Cells

LUMIC generates high-quality cell images using diffusion models and self-supervised learning, outperforming existing methods across cell lines and treatments.