Cell Image Analysis Techniques

FiberAI automates DNA-fiber analysis in microscopy images, achieving high accuracy in detection and segmentation, and enables efficient measurement of replication checkpoint integrity.

The paper introduces CLIJ-assistant, a tool that enables interactive GPU-accelerated image processing and cross-platform code generation for bio-image analysis.

The paper develops a fast image classification method for biological microscopy by selecting discriminant features and using machine learning to improve speed and accuracy.

This paper introduces a tool for segmenting and tracking lineage in noisy, low frame rate 4D fluorescence microscopy data from zebrafish embryos.

SpecSeg segments neurons and neurites in chronic calcium imaging data using cross-spectral power analysis for ROI detection and registration.

Flood-filling networks with local section realignment segment a forty-teravoxel whole-brain Drosophila electron microscopy volume, making circuit reconstruction an order of magnitude faster than manual tracing.

The paper creates a deep collection of quantitative nuclear division dynamics data from RNAi-treated C. elegans embryos, identifies 8660 reproducible RNAi phenotypes, and suggests 24 functional processes in early embryogenesis.

SAM is a MATLAB-based tool for segmenting and analyzing mother machine data to detect and correct imaging aberrations in rod-shaped bacteria.

Complete morphologies of 1,708 neurons, sparsely labelled and imaged whole-brain with fMOST, expose region-specific long-range projection rules and finer diversity than current transcriptomic subtypes capture.

The paper introduces a computational pipeline for analyzing large-scale time-lapse microscopy data to predict cancer drug response in patient-specific contexts.

SpineS is an open-source software for automatic longitudinal analysis of dendritic spine dynamics with manual intervention, showing high performance on two-photon and confocal imaging datasets.

NuMorph quantifies nuclei and nuclear markers in cleared mouse brain images, revealing gene deletion effects on cell-type distribution and brain morphology in two mouse models.

CUBIC-Cloud is a cloud-based framework for analyzing and integrating whole mouse brain data, demonstrating applications in cell distribution, plaque quantification, neuronal activity, and connectivity mapping.

SCOUT improves neuronal tracking in longitudinal calcium imaging by reducing motion artifacts and false discoveries, enabling analysis of neural ensemble dynamics over 60 days.

MORPHIOUS uses unsupervised machine learning to detect activated microglia and astrocytes by comparing to non-activated glia, identifying novel activation features.

The paper develops an automated image processing pipeline with machine learning to classify motor neuron phenotypes in ALS tissue at single-cell resolution, revealing common morphological aberrations and validating their discriminative power between ALS and healthy tissue.

This paper introduces a pipeline for 3D analysis of human brain cytoarchitecture at micrometer resolution, enabling quantitative evaluation of neuronal density and volume across different brain regions and subjects.

HistoFlow enables fast, label-efficient training of deep learning models for cell segmentation and analysis using interactive tools and artificial data.

The paper introduces a method to align images from different imaging sessions by performing immunocytochemistry on the microscope stage and using biological structures as landmarks.

EMC2 tracks calcium dynamics in neurons of behaving animals, showing daily turnover in mouse visual cortex and identifying three neuron ensembles in Hydra.

This paper introduces SRSNet, a super-resolution segmentation network that accurately reconstructs packed neurites, outperforming existing methods.

This paper introduces a new statistical tool to identify spatial patterns in histological images, applied to murine bone marrow to reveal micro-niche structures in the hematopoietic stem cell microenvironment.

flowCut is an R package that automatically detects and removes outlier events in flow cytometry data, outperforming existing methods in evaluation.

CDeep3M-Preview allows users to test pre-trained neural networks on their data or public datasets via an online tool to improve reproducibility and biological insights.

The paper introduces EASE, a method that matches calcium imaging signals to EM-derived cell segments in mouse visual cortex, enabling dendritic functional analysis and denoising.

Laocoön is a Python tool for automated cell counting in images, providing cell cycle proportions and is available on GitHub and PyPi.

BFNet visualizes uncertainty in predicting fluorescent structures from brightfield images using Monte Carlo dropout during inference.

Brainrender is a Python package for interactive visualization of anatomically registered brain data, enabling seamless integration of diverse datasets and atlases.

This paper introduces semi-automated strategies combining 3D visualization and machine guidance to accelerate EM connectome reconstruction, showing improved performance over prior methods.

neo4j-based graph data model, web explorer and HTTP/Python API for EM connectomes; on an unpublished dataset with 10s of millions of synaptic elements, most queries finish under a second.