Cell Image Analysis Techniques

The paper develops a semi-automated image-analysis pipeline to quantify myonuclear morphology and spatial organization in skeletal muscle fibers, improving efficiency and reducing bias.

SynAPSeg provides a framework and dataset for deep learning-based synapse detection and quantification, achieving expert-level performance and revealing age-related synaptic changes.

UDIST separates shape and texture in microscopy images into independent latent subspaces, enabling unbiased phenotypic analysis across biological scales.

The paper introduces two 3D neural networks for segmenting and scoring epithelial cells in cleared mammary tissue, demonstrating their use in quantifying myoepithelial cell proportions during pregnancy-lactation transitions.

This paper introduces MMB, a framework that encodes neuronal morphology into symbolic representations, revealing region-specific and scale-dependent principles of neuronal organization across the brain.

The paper attempts to reproduce findings from Bingham et al. on presynaptic actin structures using bioimage analysis, highlighting reproducibility challenges and the need for transparent methods.

The paper introduces spatialproteomics, a Python toolbox for analyzing highly multiplexed fluorescence images, demonstrating its use on lymph node and lymphoma data.

TWARDIS automates C. elegans image analysis using AI models, improving accuracy and reducing manual effort in morphometry, behavior, and neural activity studies.

This paper introduces a shape analysis algorithm, LCPC, to quantify spatial morphology and context in 2D and 3D cell cultures, capturing features like chirality that traditional metrics miss.

NeuroFlow provides a browser-based workflow for registering and quantifying mouse brain histology without requiring additional software installation or remote server uploads.

This paper introduces a deep attention network model to infer cell-cell interaction dynamics from trajectory data, providing interpretable insights into how interactions affect cell movement.

NemaSize is an AI pipeline that accurately measures nematode body size across complex postures, outperforming CellProfiler in curved and overlapping animals.

OpticalFlow3D measures amorphous motion in 3D fluorescence microscopy images without requiring segmentation or prior motion definitions.

HiExM enables scalable single-organelle analysis by improving spatial resolution and throughput, allowing detailed profiling of late endosomes and lysosomes.

The paper introduces cellmig, a Bayesian tool for quantifying cell migration velocity with uncertainty, improving reproducibility and biological insight in high-throughput assays.

The paper uses ML to analyze morphological dynamics of hiPSC-derived endothelial cells under shear stress, identifying a bifurcation between two stable cell states.

This paper introduces a noise-adaptive attention mechanism for microtubule segmentation in microscopy images, showing improved performance on synthetic and real datasets.

The paper develops a machine learning framework to automatically quantify axonal and myelin features in electron microscopy images of gene-edited mice, showing correlations between myelin changes and mitochondrial density.

This paper introduces DSC, a metric to quantify cancer cell morphology-migration coupling in live label-free imaging, showing it captures non-genetic phenotypic plasticity.

Celldetective is an AI-enhanced image analysis tool for analyzing dynamic cell interactions in immunology and immunotherapy assays.

This paper develops a machine learning method to quantify bacterial deformation and motility using convolutional neural networks, applicable to diverse transformable cells.

SpotMAX is an AI-driven framework for automated spot detection and quantification in microscopy data, outperforming existing tools and human annotators across diverse biological applications.

Stars2Cells tracks neurons across imaging sessions using a novel descriptor-based method, achieving high accuracy in identifying single neurons over time.

SImBA-SiQuAl is an open-source workflow for high-throughput quantitative analysis of 3D microtumours, enabling phenotypic profiling of spheroids and organoids.

This study assesses the reproducibility of a 3D spheroid-based assay across four institutes using dual-view oblique plane microscopy and finds high consistency in treatment effects with low technical noise.

This study introduces a semi-automated method to segment and analyze glomerular architecture from 3D confocal microscopy data, reducing manual effort and improving standardization.

This paper develops a self-supervised framework for learning structure-connectivity representations in dense connectomes, revealing neuronal organization beyond canonical labels.

The paper introduces a method to estimate proofreading progress in connectomic datasets using global graph invariants, enabling optimal stopping points for reconstruction quality.

MCD Stitcher converts proprietary .mcd IMC files into OME-TIFF images with automated stitching for whole-slide spatial analysis.