Cell Image Analysis Techniques

The paper improves image-based profiling by fusing data to capture cell heterogeneity, enhancing prediction of compound mechanisms of action and gene pathways by ~20%.

This paper develops a deep learning method to quantify oligodendrocyte ensheathment in vitro, achieving accuracy comparable to human experts and enabling high-throughput analysis.

DSeg segments and analyzes time-series images of filamentous structures, providing statistical metrics like persistence length and growth rate.

The paper develops and evaluates neural network filters to improve 3D optical microscopy images of neurites, showing significant improvements in image quality metrics compared to conventional filters.

MOSES quantifies cellular motion phenotypes across different epithelial cell types and identifies boundary formation changes in response to external stimuli.

The paper develops nTracer, an ImageJ plugin for reconstructing neuronal morphology in densely labeled mouse brains using multispectral tracing.

The paper develops an automated method to map FISH signals in sperm nuclei and finds conserved chromosome territory organization across mouse lineages.

This study uses transfer learning with pre-trained deep convolutional neural networks to classify cellular morphological changes from drug-treated cells, achieving 95-97% accuracy.

CaImAn, an open-source library, automates motion correction, source extraction, deconvolution and cross-session registration for calcium imaging, detecting neurons near human level (F1 0.75 batch, 0.76 online).

NeuroInfo automatically registers experimental mouse brain sections with a 3D atlas to accurately identify brain regions, improving navigation and reducing observer error.

LimeSeg is a 3D image segmentation method that simulates lipid membranes to segment biological structures from various images.

The paper reconstructs 1,000 projection neurons in the mouse brain, revealing new cell types and organizational principles of long-range connectivity.

The paper presents a deep learning method using GANs and feature transfer to classify cell shapes with minimal training samples, achieving high accuracy in detecting multicellular rosettes in C. elegans embryos.

Usiigaci segments and tracks cells in phase contrast microscopy using machine learning, providing accurate cell movement and morphological data for migration analysis.

The paper introduces an algorithm for analyzing filament motility with improved time resolution, showing better accuracy in detecting velocity and motile fraction changes compared to existing methods.

This paper introduces an adaptive particle representation for fluorescence microscopy images to overcome storage and processing bottlenecks by replacing pixels with content-aware particles.

RefCell analyzes image-based HTS data by identifying typical cells in reference states to assess deviations from typical behavior in perturbations or samples.

The paper uses computational modeling to analyze ommatidia formation in the Drosophila eye imaginal disc, identifying spatial patterns and spatiotemporal features.

The paper introduces an automated method to classify synaptic vesicles in C. elegans electron tomograms using machine learning, finding increased DCVs and greater distance from AZs in dauer larvae.

cytometree is an automated flow cytometry gating algorithm that outperforms existing methods in accuracy and speed.

The paper develops a model-based method to automate fecundity estimation in Drosophila by counting eggs from images, achieving a high correlation with manual counts.

This paper introduces a segmentation scheme for complex neuronal arbors, demonstrating its effectiveness on honeybee neurons.

The paper develops a 3D morphometric analysis pipeline for cell nuclei and nucleoli, achieving high classification accuracy in prostate cancer and fibroblast cells.

ACSAT automatically segments neurons in large-scale calcium imaging datasets using adaptive thresholding, achieving 80% precision and recall comparable to human inspection.

DeepMQ quantifies myelin sheaths in fluorescence images of mouse stem cell-derived cells with 93.38% accuracy using a feature extraction and deep learning classification approach.

This paper develops a deep convolutional neural network for automated astrocyte detection in microscopy images, achieving accuracy comparable to human experts and faster runtime than existing methods.

This paper introduces an experimental-computational pipeline to quantify cancer cell migration, showing that blocking sAGR2 reduces migration of MDA-MB-231 cells.

This paper introduces 3D SPECS, a method to map spatial protein localization in 3D cell cultures and analyze mitotic protein networks, revealing novel insights into spindle fragility and chromatin stress.

The paper introduces a framework for characterizing and screening antibodies against synaptic molecules for array tomography, providing objective metrics to evaluate antibody performance.

This paper introduces flood-filling networks for automated neuron reconstruction, achieving high precision in tracing neurites from electron microscopy data.