Cell Image Analysis Techniques

GraFT isolates fluorescing components in neural imaging by learning time traces with spatial maps as coefficients, using a graph filtering model based on temporal activity rather than spatial proximity.

This study applies Mask R-CNN to analyze Haemonchus contortus motility videos, showing improved detection and prediction compared to existing methods.

CellSium generates synthetic microcolony images for training deep learning segmentation models in microfluidic live-cell imaging.

SHAPR predicts 3D cell shapes from 2D images, improving accuracy over existing methods and enabling morphometric analysis of cell aggregates.

DeepSpot enhances RNA spots in smFISH images using deep learning, enabling automated detection without manual parameter tuning and achieving over 97% accuracy.

This paper develops a method to assess cancer organoid viability using rapid imaging and a fragmentation algorithm, finding significant correlations between SSD scores and traditional viability markers.

This paper introduces UPSIDE, a deep learning platform that identifies dynamic cell phenotypic states from transmitted light movies by clustering latent representations and inferring transition rates.

FlyWire enables collaborative proofreading of whole-brain connectomes through an online platform, demonstrating its utility in reconstructing and analyzing mechanosensory neuron circuits.

BioProfiling.jl provides a Julia package for morphological profiling of high-content imaging data, enabling analysis of cell state changes and integration with molecular target networks.

The paper introduces a cloud-based platform for analyzing calcium activity in motor neurons to identify compounds that reverse disease phenotypes in amyotrophic lateral sclerosis.

This paper benchmarks axon length estimation tools using simulated and real data to evaluate accuracy, efficiency, and practicality for different research contexts.

ELEPHANT enables accurate 3D cell lineage tracking in crustacean leg regeneration using an incremental deep learning approach with minimal user effort.

OnePetri uses computer vision to rapidly and accurately count bacteriophage plaques on Petri dishes, outperforming manual counting and existing tools in speed and accuracy.

MCMICRO processes multiplexed tissue images into single-cell data, demonstrating its versatility across different tissues and imaging platforms.

The paper discusses computational challenges in analyzing multiplexed imaging data and outlines community-driven approaches to address them through workshops and hackathons.

The paper introduces Mesmer, a deep learning algorithm for cell segmentation in tissue images, achieving human-level performance and enabling automated feature extraction.

Mesmerize is a user-friendly platform for analyzing 2D and 3D calcium imaging data, offering a FAIR system for processing and visualization across multiple organisms.

The paper summarizes a survey of 484 bioimage analysis community members to identify needs for tools, training, and resources.

3dSpAn is a software tool for 3D segmentation and morphological analysis of dendritic spines, addressing segmentation and quantification challenges using multiscale algorithms and feature extraction.

Origami computes oriented cell shape features in curved epithelia using fluorescence microscopy images, accurately identifying cell orientation in zebrafish embryos.

The paper presents an automated method using QuPath to quantify microglia and pericytes in whole mouse brain sections, showing regional differences in cell proportions.

The paper introduces SynapseJ, an ImageJ macro for automated synapse identification in multichannel confocal images, demonstrating its application to identify inhibitory synapses and specific neuronal connections.

This study uses a deep learning tool to analyze axonal degeneration patterns in cortical neurons, revealing morphological heterogeneity and progression dynamics.

Dice-XMBD improves cell segmentation accuracy for imaging mass cytometry data compared to existing methods.

The paper develops a deep-learning model to convert phase-contrast images into fluorescence-like images for better analysis of bacterial aggregation and rippling behaviors in Myxococcus xanthus.

This paper introduces a graph-based cell tracking algorithm with few parameters and automatic segmentation error correction, showing strong performance on degraded segmentation data.

The paper develops a U-Net-based method for cell segmentation in quantitative phase microscopy images, improves transferability with post-processing and self-supervised pretraining, and introduces a new dataset.

NeuriteNet is a CNN that sorts neuronal images into control and treatment groups based on morphological differences and highlights distinguishing features via saliency maps.

This paper introduces a machine learning method for cell segmentation in developing embryos, showing improved accuracy over existing techniques.