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 Experiment Video

Updated: Jun 12, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Leveraging data-driven segmentation uncertainty estimates as potential diagnostic indicators.

Garrett Regan1, Joel G Fletcher2, David R Holmes1

  • 1Department of Biomedical Engineering and Physiology, Mayo Clinic, Rochester, MN, USA.

Quantitative Imaging in Medicine and Surgery
|June 11, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Uncertainty: Confidence Intervals00:54

Uncertainty: Confidence Intervals

The confidence interval is the range of values around the mean that contains the true mean. It is expressed as a probability percentage. The interpretation of a 95% confidence interval, for instance, is that the statistician is 95% confident that the true mean falls within the interval. The upper and lower limits of this range are known as confidence limits. The confidence limits for the true mean are estimated from the sample's mean, the standard deviation, and the statistical factor 't,' or...

You might also read

Related Articles

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

Sort by
Same author

Fundamental concepts of effective troponin use: important principles for internists.

The American journal of medicine·2014
Same author

A prospective randomized controlled study of erythromycin on gastric and small intestinal distention: implications for MR enterography.

European journal of radiology·2014
Same author

Delayed transcatheter heart valve migration and failure.

JACC. Cardiovascular imaging·2014
Same author

Aortic valve replacement for severe aortic valve stenosis in the nonagenarian patient.

The Annals of thoracic surgery·2014
Same author

Snake eyes.

Journal of the American College of Cardiology·2014
Same author

Novel balloon catheter device with pacing, ablating, electroporation, and drug-eluting capabilities for atrial fibrillation treatment--preliminary efficacy and safety studies in a canine model.

Translational research : the journal of laboratory and clinical medicine·2014
Same journal

Amide proton transfer-weighted magnetic resonance imaging for predicting histopathology and biomarkers in rectal adenocarcinoma.

Quantitative imaging in medicine and surgery·2026
Same journal

Multimodality imaging for diagnosing and monitoring immunoglobulin G4-related coronary arteritis presenting as giant aneurysm: a case description.

Quantitative imaging in medicine and surgery·2026
Same journal

Investigation of the topological properties of brain structural and functional networks in patients with mild cognitive impairment.

Quantitative imaging in medicine and surgery·2026
Same journal

The critical role of transesophageal echocardiography in diagnosing carbon dioxide gas embolism: a case description and lessons learned.

Quantitative imaging in medicine and surgery·2026
Same journal

Impact of data augmentation size on deep learning-based third lumbar vertebra computed tomography skeletal muscle segmentation performance.

Quantitative imaging in medicine and surgery·2026
Same journal

Quantitative measurement of cutaneous neurofibromas in neurofibromatosis type 1 using a structured-light scanner.

Quantitative imaging in medicine and surgery·2026
See all related articles

Quantifying uncertainty in AI segmentation models can identify potential pathology. This study shows that uncertainty estimates are higher in diseased tissues, aiding in early detection across multiple organs.

Area of Science:

  • Artificial intelligence in medical imaging
  • Medical image segmentation
  • Explainable AI

Background:

  • Quantifying uncertainty is crucial for explainable AI, especially in clinical settings.
  • Current research on uncertainty estimates in segmentation models is limited to simple, single-class problems.
  • This study aims to extend uncertainty analysis to complex segmentation tasks and explore its potential as a pathology marker.

Purpose of the Study:

  • To characterize abdominal segmentation uncertainty using a Multi-Organ Test-Time Augmentation (TTA) pipeline.
  • To evaluate the feasibility of using uncertainty information as a marker for potential pathology in abdominal organs.
  • To expand uncertainty analysis to complex, multi-organ segmentation problems.

Main Methods:

Keywords:
SegmentationTest-Time Augmentation (TTA)uncertainty

Related Experiment Videos

Last Updated: Jun 12, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

  • Utilized the TotalSegmentator model and a custom Multi-Organ Test-Time Augmentation (TTA) pipeline.
  • Characterized uncertainty across 14 organs and 8 augmentations in 872 abdominal CT cases.
  • Quantified uncertainty by averaging entropy within dilated organ masks and evaluated pathology detection in 489 additional cases.

    • Main Results:

    • Uncertainty estimates were significantly higher in pathologic tissues compared to healthy tissues across all evaluated organs (P < 1.79e-05).
    • Receiver operating characteristic analysis demonstrated pathology detection performance ranging from fair (pancreas, colon, left kidney) to excellent (liver).
    • Area under the curve (AUC) values for pathology detection ranged from 0.7097 to 0.8885, indicating robust performance.

    Conclusions:

    • Test-Time Augmentation (TTA)-derived uncertainty estimates are valuable tools for detecting pathology across multiple organ systems.
    • Future development could lead to tools for early pathology screening or active learning in medical image analysis.
    • This approach offers a promising method for identifying regions requiring manual intervention or further investigation.