Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Changes in Retinal Microvasculature During Healthy Pregnancy Measured by AO.

Hypertension (Dallas, Tex. : 1979)·2026

Same author

C-terminal long-QT type 1 R562S-Kv7.1 variant, the first variant in helix C impairing β-adrenergic response of the slow delayed rectifier K+ channel.

Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology·2026

Same author

Automated attenuation analysis of CT pulmonary angiography identifies peripheral hyperperfusion as a prognostic marker in non-surgical chronic thromboembolic pulmonary hypertension (CTEPH).

Quantitative imaging in medicine and surgery·2026

Same author

Basecalling-free resistance gene identification using a hybrid transformer in raw nanopore signals.

Frontiers in microbiology·2026

Same author

Non-Contrast Radiation-Free NIR Lung Imaging.

Biomedicines·2025

Same author

Ropivacaine and lidocaine inhibit SaOS-2 osteosarcoma cell proliferation and invasion via mitochondrial dysfunction and apoptosis.

European journal of cell biology·2025

Search research articles

Related Experiment Video

Updated: Oct 14, 2025

Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning

Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning

Published on: November 19, 2018

Self-supervised pretraining for transferable quantitative phase image cell segmentation.

Tomas Vicar^1,2, Jiri Chmelik¹, Roman Jakubicek¹

¹Department of Biomedical Engineering, Faculty of Electrical Engineering and Communication, Brno University of Technology, Brno, Czech Republic.

Biomedical Optics Express

|November 8, 2021

Summary

This summary is machine-generated.

A new U-Net based method enhances adherent cell segmentation in quantitative phase microscopy. Self-supervised pretraining and adjustable post-processing improve accuracy for diverse cell types.

More Related Videos

Discrimination and Characterization of Heterocellular Populations Using Quantitative Imaging Techniques

Discrimination and Characterization of Heterocellular Populations Using Quantitative Imaging Techniques

Published on: June 30, 2017

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Related Experiment Videos

Last Updated: Oct 14, 2025

Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning

Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning

Published on: November 19, 2018

Discrimination and Characterization of Heterocellular Populations Using Quantitative Imaging Techniques

Discrimination and Characterization of Heterocellular Populations Using Quantitative Imaging Techniques

Published on: June 30, 2017

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Area of Science:

* Biomedical imaging
* Computational biology
* Machine learning

Background:

* Accurate segmentation of adherent cells is crucial for quantitative analysis in microscopy.
* Existing methods may struggle with variations in cell morphology and imaging conditions.
* Robust segmentation is essential for reliable cell-based research and diagnostics.

Purpose of the Study:

* To develop and optimize a novel U-Net based deep learning method for robust adherent cell segmentation.
* To enhance the transferability of the segmentation model across different cell types using adjustable post-processing.
* To introduce a self-supervised pretraining strategy to improve segmentation performance using unlabeled data.

Main Methods:

* Design and implementation of a U-Net based convolutional neural network architecture.
* Evaluation of four distinct post-processing pipelines for segmentation refinement.
* Application of non-deep learning transfer with adjustable parameters for improved cross-cell type performance.
* Development and application of a self-supervised pretraining technique using image reconstruction from distortions.

Main Results:

* The proposed U-Net based method achieved robust adherent cell segmentation.
* The self-supervised pretraining improved segmentation performance, increasing the object-wise intersection over union from 0.67 to 0.70.
* Adjustable post-processing pipelines enhanced the model's transferability to different cell types.
* A new dataset of labeled and unlabeled images was created and published.

Conclusions:

* The developed U-Net based method offers a robust solution for adherent cell segmentation in quantitative phase microscopy.
* Self-supervised pretraining and adaptable post-processing significantly enhance segmentation accuracy and generalizability.
* The released dataset will facilitate further research in cell segmentation and self-supervised learning for microscopy images.