Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Structural relaxation and domain formation in anisotropically strained La<sub>0.7</sub>Sr<sub>0.3</sub>MnO<sub>3</sub>/LaFeO<sub>3</sub> superlattices on DyScO<sub>3</sub>(101).

Scientific reports·2026
Same author

Self-rated health as a predictor of mortality and healthcare use in older adults at high risk of hospitalisation: a prospective cohort study in Sweden.

BMJ open·2025
Same author

Accessible and thorough magnetic field mapping of TEMs using a commercial Hall-sensor and TEM automation for in situ magnetic experiments.

Micron (Oxford, England : 1993)·2025
Same author

Association between frailty and pain in older people at high risk of future hospitalization.

Frontiers in pain research (Lausanne, Switzerland)·2025
Same author

On the effect of precession for magnetic differential phase contrast imaging.

Micron (Oxford, England : 1993)·2024
Same author

Reasons for hospitalisation and cumulative mortality in people, 75 years or older, at high risk of hospital admission: a prospective study.

BMC geriatrics·2024
Same journal

Deep PACBED: Multitask analysis of PACBED images using deep neural networks.

Ultramicroscopy·2026
Same journal

Guided progressive reconstructive imaging: A new quantization-based framework for low-dose, high-throughput and real-time analytical ptychography.

Ultramicroscopy·2026
Same journal

Brightness optimization in a 200 keV DTEM source by geometry-driven aberration suppression.

Ultramicroscopy·2026
Same journal

Characterization of the Timepix4 hybrid pixel detector and its impact on four-dimensional scanning transmission electron microscopy (4D-STEM).

Ultramicroscopy·2026
Same journal

Contamination analysis of the residual gas composition in transmission electron microscopy.

Ultramicroscopy·2026
Same journal

Temperature-dependent mean inner potential of polystyrene spheres measured using off-axis electron holography.

Ultramicroscopy·2026
See all related articles

Related Experiment Video

Updated: Jun 10, 2025

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.6K

Exploring deep learning models for 4D-STEM-DPC data processing.

Gregory Nordahl1, Sivert Dagenborg1, Jørgen Sørhaug1

  • 1Department of Physics, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway.

Ultramicroscopy
|October 10, 2024
PubMed
Summary
This summary is machine-generated.

Supervised learning with convolutional neural networks (CNNs) offers automated processing for nanoscale magnetic material analysis using differential phase contrast (DPC) imaging. This approach enhances consistency and accuracy in analyzing four-dimensional scanning transmission electron microscopy (4D-STEM) data.

Keywords:
4D-STEMConvolutional neural networksDifferential phase contrastMagnetic thin film imaging

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Three and Four-Dimensional Visualization and Analysis Approaches to Study Vertebrate Axial Elongation and Segmentation
12:59

Three and Four-Dimensional Visualization and Analysis Approaches to Study Vertebrate Axial Elongation and Segmentation

Published on: February 28, 2021

3.7K

Related Experiment Videos

Last Updated: Jun 10, 2025

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.6K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Three and Four-Dimensional Visualization and Analysis Approaches to Study Vertebrate Axial Elongation and Segmentation
12:59

Three and Four-Dimensional Visualization and Analysis Approaches to Study Vertebrate Axial Elongation and Segmentation

Published on: February 28, 2021

3.7K

Area of Science:

  • Materials Science
  • Nanoscience
  • Data Science

Background:

  • Differential phase contrast (DPC) imaging is crucial for nanoscale magnetic materials.
  • Advancements in direct detector technology drive the need for enhanced four-dimensional scanning transmission electron microscopy (4D-STEM) data processing.
  • Conventional DPC processing algorithms are often tailored to specific experimental conditions.

Purpose of the Study:

  • To explore the application of supervised learning with convolutional neural networks (CNNs) for automated STEM-DPC data processing.
  • To investigate two distinct CNN-based approaches for DPC data analysis.
  • To compare the performance of CNN models against conventional algorithms, especially concerning diffraction contrast and domain wall characterization.

Main Methods:

  • Training CNNs on experimentally acquired 4D-STEM data.
  • Implementing a regression-based approach for direct beam tracking.
  • Utilizing a modified U-net architecture for direct beam segmentation as a pre-processing step.
  • Comparing model outputs with conventional algorithms for accuracy and consistency.

Main Results:

  • CNNs effectively process STEM-DPC data acquired under similar instrument parameters.
  • The developed models demonstrate robust performance even in the presence of strong diffraction contrast.
  • Analysis of domain wall profiles and width measurements shows comparable or improved results using CNNs.

Conclusions:

  • Supervised learning with CNNs provides a powerful and automated solution for STEM-DPC data processing.
  • These methods enhance consistency and reliability in nanoscale magnetic material characterization.
  • CNN-based approaches offer a viable alternative to conventional, experimentally tailored algorithms.