Cell Image Analysis Techniques

CGMD: Centrality-Guided Maximum Diversity for Annotation-Efficient Fine-Tuning of Pretrained Cell Segmentation Models

CGMD selects a small set of diverse images for fine-tuning pretrained cell segmentation models, achieving performance close to fully supervised methods with minimal annotations.

Real-Time Multi-Position and Multi-ROI Tracking with LiLiTTool for Smart Light-Sheet Microscopy in Growing Samples

LiLiTTool enables real-time tracking of multiple regions of interest in growing embryos using deep learning and sensor fusion, improving long-term imaging reliability.

Automated Proofreading of Digitally Reconstructed NeuralMorphology Enhances Accuracy, Scalability, and Standardization

The paper presents an automated machine learning pipeline for standardizing and correcting neural morphology reconstructions, achieving high accuracy in dendritic relabeling and structural normalization.

SegBio: A lightweight end-to-end toolkit for Instance Segmentation of biological samples

SegBio provides a lightweight, user-friendly pipeline for instance segmentation of crowded biological images, achieving high performance on C. elegans datasets.

NucleoNet and DropNet: Generalist deep learning models for instance segmentation of nuclei and lipid droplets from electron microscopy images

The paper develops NucleoNet and DropNet, deep learning models for segmenting nuclei and lipid droplets in EM images, and applies them to compare 2D/3D cancer models with in vivo tumors.

A quantitative census of millions of postsynaptic structures in a large electron microscopy volume of mouse visual cortex

The paper presents a computational pipeline for classifying postsynaptic targets in neuronal meshes, achieving high accuracy on mouse visual cortex data and demonstrating generalizability to other connectomes.

DNAi: an open-source AI tool for unbiased DNA fiber analysis

DNAi is an open-source deep learning tool that automates DNA fiber analysis, matching human performance in segmentation and length measurement.

Object Detection Techniques for Live Monitoring of Amoeba in Phase-Contrast Microscopic Images

This study develops and evaluates deep-learning models for amoeba detection in phase-contrast images using annotated datasets and compares their speed and accuracy for real-time applications.

Scalable data harmonization for single-cell image-based profiling with CytoTable

The paper introduces CytoTable, a software package that harmonizes single-cell image-based profiling data from different tools, enabling reproducible and scalable data integration.

An Open-Source Code To Analyze Mitochondrial Intracellular Distribution From Fluorescence Microscopy Images

The paper presents an open-source code to estimate mitochondrial radial localization from fluorescence microscopy images using various markers.

NetTracer3D Enables User-Friendly Analysis of Diverse Microscopic and Medical 3D Datasets

NetTracer3D provides a tool for analyzing 3D tissue maps from diverse datasets, enabling network-based exploration of spatial relationships and tissue organization.

Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes

Cell Painting in primary human hepatocytes exposed to 1,085 compounds detects bioactivity at lower concentrations than cytotoxicity assays and predicts cell-based, but not cell-free, ToxCast endpoints across feature extractors.

Software for semi-automatic analysis of microscopic images of adhesion structures and protein colocalization in cells

The paper introduces a semi-automatic MATLAB script for analyzing adhesion structures and protein colocalization in microscopic images, improving efficiency and reducing subjectivity.

SynThIA: A semi-automated tool for quantification of multi-partite synapses

SynThIA is a semi-automated tool for quantifying multi-partite synapses using up to four markers, offering high-throughput and accurate analysis.

MORPHIS (MORPHological Interpretable Signature) captures heterogeneous treatment- and aging-related responses of single cells

MORPHIS quantifies single-cell morphological responses to treatments and aging, distinguishing compound-specific signatures and aging-related changes in Caco-2, HeLa, and C. elegans cells.

Spoti-find: A novel, open-source void spot assay image analysis tool

Spoti-find is an open-source tool for analyzing void spot assays, introducing new metrics and improving usability and accuracy in urinary behavior studies.

Random-forest segmentation and spatial analysis of injected cardiac spheroids in optically cleared myocardium

This paper develops a random forest-based method to segment and analyze transplanted cardiac spheroids in optically cleared myocardium, achieving accurate 3D quantification.

Quantitative 3D cytoarchitecture of human brain organoids using light-sheet microscopy

The paper introduces LUCID-org, a method for 3D imaging and analysis of human brain organoids, revealing cytoarchitectural defects in CENPJ-mutated microcephaly organoids.

CellPheno: A High-throughput Computational Platform for Quantifying Cellular Resolution Whole Brain Microscopy Images

CellPheno is a 3D nuclei segmentation tool for whole-brain microscopy that enables efficient morphometry and co-localization analysis across multiple cell types in mouse brains.

Learning Continuous Morphological Trajectories via Latent Principal Curves

The paper introduces MorphCurveVAE, a method to infer continuous morphological trajectories from static 3D microscopy images, demonstrating its effectiveness on mitosis data.

Establishment of an optimized and automated workflow for whole brain probing of neuronal activity.

The paper develops an automated workflow for whole brain analysis of neuronal activity using c-Fos immunohistochemistry and open-source tools, enabling accurate and unbiased quantification across contexts.

Calcium transient detection and segmentation with the astronomically motivated algorithm for background estimation and transient segmentation (Astro-BEATS)

Astro-BEATS detects and segments miniature synaptic calcium transients in fluorescence imaging using astronomical transient detection techniques, outperforming threshold-based methods and enabling efficient training of deep learning models.

Universal high-throughput image quantification of subcellular structure dynamics and spatial distributions within cells

This paper introduces GRID, a high-throughput image analysis method using dispersion indices to quantify subcellular structure dynamics and spatial distributions across diverse biological contexts.

SAMWOOD: An automated method to measure wood cells along growth orientation

SAMWOOD automates wood cell segmentation using a zero-shot model, enabling precise anatomical measurements and reducing analysis time and bias.

mnDINO: Accurate and robust segmentation of micronuclei with vision transformer networks

mnDINO is a vision transformer model for accurate micronuclei segmentation in DNA-stained images, achieving high accuracy and generalization across cell lines and microscopes.

BrAIn: A comprehensive artificial intelligence-based morphology analysis system for brain organoids and neuroscience

BrAIn quantifies brain organoid morphology and evaluates growth conditions' effects on development and complexity.

Accelerated Discovery of Cell Migration Regulators Using Label-Free Deep Learning-Based Automated Tracking

This paper introduces DeepBIT, a deep learning method for automated cell migration tracking, enabling high-throughput analysis of cancer cell motility under various conditions and identifying new regulators of migration.

AQuA2-Cloud: a web platform for fluorescence bioimaging activity analysis

AQuA2-Cloud provides a web platform for analyzing fluorescence bioimaging data, enabling cloud-based activity quantification with improved accessibility and usability compared to the original AQuA2 tool.

imAgeScore, a Cell Painting-Based Predictor of Cellular Age for High-throughput Drug Screening Applications

imAgeScore predicts cellular age using Cell Painting features, identifies age-modulating compounds, and validates rejuvenation effects in human fibroblasts.

NeuronID: An automatic toolkit for identifying neurons in two-photon calcium imaging data

NeuronID is a toolkit for automated processing of two-photon calcium imaging data, reducing over-segmentation and achieving near-human accuracy.