Cell Image Analysis Techniques

DELiVR uses VR-generated data to train deep learning models for quantifying c-Fos in cleared mouse brains, revealing new activation patterns in cancer-related brain activity.

This paper introduces asymmetric contrastive objectives for phenotypic screening, improving performance on benchmark datasets and real-world cell screening tasks.

This study compares drug toxicity measurements using XTT assay, quantitative nuclei imaging, and trypan blue exclusion, showing imaging provides higher accuracy and better consistency with drug mechanisms.

The paper introduces a new image analysis pipeline for mother machine data and evaluates its performance against existing tools, highlighting parameter sensitivity and segmentation biases.

Acto3D provides an open-source, user-friendly software for generating and analyzing high-resolution 3D images with multiple fluorescence channels, validated on mouse embryos.

This paper proposes a customized CycleGAN to enhance cell segmentation with limited annotated data, showing improved performance over conventional methods.

RASP detects fluorescent puncta in complex cellular backgrounds with high precision and speed, outperforming existing methods by using image gradients to separate objects from noise.

The paper introduces a dataset of cell morphology images treated with matched chemical and genetic perturbations and evaluates methods for identifying perturbation similarities.

The paper benchmarks cellular segmentation methods, showing Cellpose outperforms Mediar when tested fairly and a transformer-based variant of Cellpose does not improve performance.

This paper introduces a new dataset for cell detection in microscopy images using oriented bounding boxes and evaluates its effectiveness for cancer biology tasks.

This paper introduces flow zoometry, a high-throughput screening method for Drosophila, enabling rapid, automated, and high-resolution analysis of whole animals.

The paper introduces a MATLAB tool for quantitatively analyzing 2D tissue images, measuring morphology and co-localization parameters with flexibility across different tissue types.

This paper improves axon segmentation in light-sheet microscopy images by fine-tuning the TrailMap model with expert-labeled data, achieving better recall and adjusted F1-score.

SC-Track improves single-cell lineage tracking by reducing noise in cell segmentations and classifications from CNNs, outperforming existing methods across diverse segmentation types.

Cellects quantifies cell expansion, motion, and morphology from 2D image data and time-lapse sequences, providing standardized outputs for statistical analysis.

This paper uses deep learning to analyze bacterial communities in soil chips from three locations, quantifying abundance and morphology to study population dynamics and biodiversity.

NIEND enhances neuronal images to improve reconstruction quality and efficiency, achieving higher F1 scores and faster processing times.

RoboEM automates 3D flight tracing for synaptic-resolution connectomics, reducing computational annotation costs for cortical connectomes by about 400-fold compared to manual error correction.

NeuronBridge enables rapid morphological matching of neurons across large datasets imaged with different modalities.

Enso decodes bacterial cell cycle and cell type from static SMLM images using unsupervised deep learning without prior knowledge.

The paper develops an imaging assay with CNNs to accurately identify cell types in dense mixed neural cultures, outperforming time-based classification and enabling quality control of iPSC-derived models.

PIFiA is a self-supervised method for protein functional annotation from single-cell imaging data, outperforming existing approaches and enabling downstream analysis of protein organization.

This paper proposes an RNN-based method to automatically detect mitosis stages from cell division videos, showing improved performance over traditional classifiers.

This study compares temporal deep learning models for classifying sequences of cell colony images to understand cellular state transitions.

The paper identifies two EB distribution patterns in Drosophila midguts and shows how Septate Junction proteins affect EB morphology and clustering during regeneration.

U-FISH improves FISH spot detection accuracy and generalization using a deep learning approach with a U-Net model and a large annotated dataset.

DeepD3 provides an automated framework for quantifying dendritic spines from microscopy data, validated across multiple datasets and conditions.

This paper develops a machine learning approach to classify 53BP1 foci in 2D images, achieving high performance with standard classifiers and a diverse parameter space.

The paper presents an image analysis pipeline for detecting circulating hybrid cells in immunofluorescence images with human-level accuracy using a β-VAE and SVM classifier.

The paper develops a weakly supervised learning method to learn treatment effect representations from imaging data, improving downstream analysis performance.