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

Identification of a sequence element regulating H3K9 methylation at the ap2-g locus in Plasmodium falciparum.

Scientific reports·2026
Same author

The role of the tryptophan-rich allosteric network and sodium egress in GPCR activation.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Molecular Design with Artificial Intelligence: Progress and Perspectives for Small Molecules.

Chemical reviews·2026
Same author

Nonnegative matrix factorization incorporating domain specific constraints for four dimensional scanning transmission electron microscopy.

Scientific reports·2025
Same author

DiffBreed: automatic differentiation enables efficient gradient-based optimization of breeding strategies.

Bioinformatics (Oxford, England)·2025
Same author

Harnessing Explainable AI to Explore Structure-Activity Relationships in Artificial Olfaction.

ACS applied materials & interfaces·2025

Related Experiment Video

Updated: Mar 16, 2026

Analysis of SEC-SAXS data via EFA deconvolution and Scatter
10:59

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Published on: January 28, 2021

10.1K

Sparse modeling of EELS and EDX spectral imaging data by nonnegative matrix factorization.

Motoki Shiga1, Kazuyoshi Tatsumi2, Shunsuke Muto2

  • 1Department of Electrical, Electronic and Computer Engineering, Gifu University, 1-1, Yanagido, Gifu 501-1193, Japan.

Ultramicroscopy
|August 17, 2016
PubMed
Summary

This study introduces a new statistical method using non-negative matrix factorization (NMF) to automatically analyze complex chemical data from scanning transmission electron microscopy (STEM) spectral imaging (SI). The approach efficiently resolves and extracts chemical components, improving data analysis accuracy.

Keywords:
Automatic relevance determinationNonnegative matrix factorizationSparse modelingSpatial orthogonalitySpectral imaging

More Related Videos

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

13.1K
Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations
12:09

Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations

Published on: January 8, 2013

14.1K

Related Experiment Videos

Last Updated: Mar 16, 2026

Analysis of SEC-SAXS data via EFA deconvolution and Scatter
10:59

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Published on: January 28, 2021

10.1K
Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

13.1K
Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations
12:09

Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations

Published on: January 8, 2013

14.1K

Area of Science:

  • Materials Science
  • Analytical Chemistry
  • Microscopy

Background:

  • Scanning transmission electron microscopy (STEM) generates large spectral imaging (SI) datasets (electron energy-loss/energy-dispersive X-ray spectroscopy).
  • Manual analysis of these datasets to identify chemical components and states is time-consuming and inefficient.
  • Statistical methods offer a more effective approach for automated data resolution.

Purpose of the Study:

  • To develop an advanced non-negative matrix factorization (NMF) model for automated chemical component extraction from STEM-SI data.
  • To optimize the number of components and enhance spectral resolution using novel penalty terms.

Main Methods:

  • Proposed a novel non-negative matrix factorization (NMF) model incorporating an automatic relevance determination (ARD) prior and a soft orthogonal constraint.
  • Developed a fast optimization algorithm based on hierarchical alternating least-squares for NMF factorization.
  • Validated the model using both phantom and real STEM-EDX/EELS SI datasets.

Main Results:

  • The ARD prior effectively determined the correct number of physically meaningful chemical components.
  • The soft orthogonal constraint demonstrated significant utility in resolving spectral components, especially for dense STEM-EELS SI data.
  • The proposed method provides efficient and accurate analysis of complex spectral datasets.

Conclusions:

  • The novel NMF model offers a robust and efficient solution for automated chemical state analysis in STEM-SI.
  • This approach significantly enhances the interpretation of complex materials characterization data.
  • The method is particularly beneficial for datasets lacking sparsity in spatial or spectral information.