Cell Image Analysis Techniques

This paper proposes an RNN-based method for automatic cell-cycle stage classification using time-lapse microscopy images, achieving superior performance over a classifier baseline.

The paper compares morphology and gene expression data to determine their complementary roles in mapping cell states across drug perturbations.

SalienceNet enhances nuclei saliency in low-light microscopy images, improving segmentation accuracy by 30% with Otsu thresholding and 26% with StarDist.

The paper introduces SCIP, a scalable image processing pipeline for morphological profiling of image cytometry data, demonstrating its application on three datasets to classify and analyze cell types and drug responses.

SOPRA is a method for analyzing high-content screens by regressing non-averaged cell feature data to identify significant changes in cell populations.

The paper improves GAN-based deep learning models for cell organization and introduces evaluation criteria to assess synthetic cell images' accuracy.

The paper introduces BEN, a method to correct batch effects in deep learning for microscopy images by aligning experimental and deep learning batches, improving model generalization.

This study compares the predictive performance of models trained on fluorescence and brightfield images for drug mechanism of action prediction, finding comparable results.

The paper introduces a deep-learning method called TPS to quantify zonal gene expression patterns in tissues like the liver, showing it can detect differences between strains, ages, and disease states.

SATINN automates testis histology analysis using CNNs to classify cell types and tubule stages in mouse mutants with high accuracy.

The paper introduces nCorrect, an algorithm that improves neuron tracing by reanalyzing existing traces to produce more physiologically-relevant reconstructions using a 4D minimization algorithm and a new similarity metric.

BioImageIT integrates FAIR data management with analysis to address challenges in bioimaging data sharing and annotation.

The paper proposes a hybrid deep learning method to exclude non-chromosome objects from metaphase chromosome images, achieving high specificity in segmentation.

H5CellProfiler is an R package that enables user-friendly analysis of large single-cell datasets on a local desktop with efficient memory usage and track labeling.

The paper introduces a method to decompose neuropil calcium imaging data into bouton-like structures and finds the resulting network is scale-free.

The paper introduces NuCLear, a method combining in vivo two-photon imaging and deep learning to monitor tissue composition over time in mice brains.

This paper introduces a polymer physics-based method to classify neurons using 3D coordinates, showing it complements existing methods and provides a physical understanding of neuronal morphology.

CardioVinci uses deep learning to automate the analysis of 3D cardiac ultrastructure changes from large EM imaging datasets.

The paper introduces a method for precise single-cell tracking and lineage reconstruction using cell region fingerprints, improving tracking performance across species and data types.

GOAT is a deep learning tool for annotating organoids from microscopy images, showing high performance across diverse organoid systems.

This paper introduces a method for automated lineage reconstruction in developing embryos using sparse annotations, achieving 75.8% accuracy on mouse embryos compared to 31.8% previously.

The paper identifies small molecule regulators of biological pathways by matching compound effects to known pathway phenotypes in public cell image data.

pcnaDeep tracks single-cells and assigns mother-daughter relationships using deep-learning and cell cycle profiling from fluorescent biosensors.

The paper develops a method to remove technical artifacts from high-throughput imaging assays using systematic replication, improving biological signal clarity.

The paper introduces a fisheye transformation for single-cell phenotyping, improving accuracy by weighting nearby cells more heavily than distant ones, and validates it on microscopy and wildlife datasets.

The paper introduces Zellige, a software tool for extracting multiple surfaces from 3D microscopy images, demonstrating its performance on synthetic and biological samples.

ASHLAR is a Python tool that stitches and registers highly multiplexed whole slide images to generate accurate mosaics, outperforming existing software.

Griottes generates network representations of segmented tissue images to extract cell interactions and properties efficiently.

The paper presents a similarity search tool for comparing neural morphologies across species and brain regions using morphometric and vectorized descriptors.

The paper uses deep generative networks to predict nuclear position from actin filament arrangements in cells, achieving over 70% accuracy in localization.