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

Evaluating Iterative Deep Learning as a Labeling-Efficient Strategy for Tubular Segmentation in Digital Nephropathology.

Journal of imaging informatics in medicine·2026
Same author

What practicing pathologists and oncologists should know about the new computational pathology-based companion diagnostic tools.

The Journal of pathology·2026
Same author

StainStyleSampler: clustering-based sampling of whole slide image appearances.

Diagnostic pathology·2026
Same author

Sacubitril/valsartan preserves kidney function in rats with cardiorenal syndrome after myocardial infarction.

PloS one·2025
Same author

Unsupervised learning for labeling global glomerulosclerosis.

Computers in biology and medicine·2025
Same author

Comparing non-machine learning vs. machine learning methods for Ki67 scoring in gastrointestinal neuroendocrine tumors.

Scientific reports·2025

Related Experiment Video

Updated: Jan 13, 2026

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

7.3K

Morphometry-based detection of deep learning faults in glomerular segmentation.

Hrafn Weishaupt1, Justinas Besusparis1, Nazanin Mola1

  • 1Department of Pathology, Haukeland University Hospital, Bergen, Norway.

Biorxiv : the Preprint Server for Biology
|January 9, 2026
PubMed
Summary

Shape analysis automatically evaluates kidney image annotations from deep learning models. This method efficiently identifies segmentation errors, allowing pathologists to focus on the most suspicious results, saving time.

Keywords:
Deep learningGlomeruliMorphometryNephropathologySegmentation evaluationShape analysis

More Related Videos

Author Spotlight: Aiding Research in Kidney Biology by Labeling Glomeruli in Cleared Tissues
09:50

Author Spotlight: Aiding Research in Kidney Biology by Labeling Glomeruli in Cleared Tissues

Published on: February 9, 2024

1.8K
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

3.3K

Related Experiment Videos

Last Updated: Jan 13, 2026

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
09:16

Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis

Published on: June 18, 2020

7.3K
Author Spotlight: Aiding Research in Kidney Biology by Labeling Glomeruli in Cleared Tissues
09:50

Author Spotlight: Aiding Research in Kidney Biology by Labeling Glomeruli in Cleared Tissues

Published on: February 9, 2024

1.8K
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

3.3K

Area of Science:

  • Nephrology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Deep learning excels at segmenting glomeruli in kidney biopsies.
  • Manual validation of AI annotations is time-consuming for pathologists.
  • Automated methods are needed to identify and correct AI segmentation errors.

Purpose of the Study:

  • To investigate shape analysis for automatically evaluating deep learning-based glomerular annotations.
  • To determine if morphometric features can detect segmentation inconsistencies.
  • To develop a strategy for prioritizing annotations for manual review.

Main Methods:

  • Extensive study of shape descriptors on over 168,000 glomerular predictions.
  • Application of morphometry to analyze segmentation inconsistencies.
  • Ranking of annotations using shape descriptors to identify errors.

Main Results:

  • Shape analysis successfully identified three types of segmentation inconsistencies.
  • Ranking annotations by shape descriptors enriched errors at the top.
  • A panel of three shape descriptors efficiently enriched errors, regardless of type.

Conclusions:

  • Shape analysis is a viable method for evaluating glomerular segmentation.
  • This approach significantly reduces the workload for pathologists by prioritizing suspicious annotations.
  • Automated error detection using shape analysis enhances the efficiency of correcting deep learning-derived annotations.