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

Indigenous underutilized fruits in Bangladesh as future resilient foods: insight into the nutritional value, bioactive compounds, and health benefits.

Critical reviews in food science and nutrition·2026
Same author

Toward Patient-Specific Partial Point Cloud to Surface Completion for Pre to Intra-operative Registration in Image-Guided Liver Interventions.

Medical Image Understanding and Analysis. Medical Image Understanding and Analysis (Conference)·2026
Same author

Evaluation of Intra-operative Patient-specific Methods for Point Cloud Completion for Minimally Invasive Liver Interventions.

Proceedings of SPIE--the International Society for Optical Engineering·2026
Same author

Investigating the Domain Adaptability of General-Purpose Foundation Models for Left Atrium Segmentation from MR Images.

Functional imaging and modeling of the heart : ... International Workshop, FIMH ..., proceedings. FIMH (Conference)·2026
Same author

Studying Perfusion Effects on Heat Transfer in Tissue-Mimicking Phantoms for Cardiac Ablation: A Preliminary Experimental Investigation.

Journal of fluid flow, heat and mass transfer·2026
Same author

Upcycling versus recycling in fast fashion: A systematic-critical review of circular economy solutions.

Waste management & research : the journal of the International Solid Wastes and Public Cleansing Association, ISWA·2026

Related Experiment Video

Updated: Sep 21, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.9K

Calibration of cine MRI segmentation probability for uncertainty estimation using a multi-task cross-task learning

S M Kamrul Hasan1,2, Cristian A Linte1,2,3

  • 1Biomedical Modeling, Visualization and Image-guided Navigation (BiMVisIGN) Lab, RIT.

Proceedings of Spie--The International Society for Optical Engineering
|May 31, 2022
PubMed
Summary

This study introduces a novel method using uncertainty estimation to detect errors in deep learning segmentation masks for cardiac MRI. This approach helps identify unreliable predictions, improving clinical applications of artificial intelligence in medical imaging.

Keywords:
Bayesian multi-task cross-task learningMonte-Carlo samplingcardiac imagingcine MR imagedeep learningerror estimationimage segmentationmyocardiumuncertaintyventricle blood-pool

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

3.0K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

692

Related Experiment Videos

Last Updated: Sep 21, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.9K
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

3.0K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

692

Area of Science:

  • Medical Image Analysis
  • Deep Learning
  • Artificial Intelligence in Healthcare

Background:

  • Deep learning (DL) models show promise in medical image analysis but lack reliability in clinical settings due to prediction failures.
  • A significant limitation of DL models is their requirement for large, labeled datasets, hindering real-world application.
  • Unreliable segmentation masks from DL models can lead to diagnostic errors and impact patient care.

Purpose of the Study:

  • To develop a novel method incorporating uncertainty estimation for detecting segmentation failures in Convolutional Neural Networks (CNNs).
  • To evaluate the correlation between uncertainty estimates and segmentation errors for a given DL model.
  • To introduce a multi-task cross-task learning consistency approach linking pixel-level segmentation and geometric-level distance map tasks.

Main Methods:

  • Proposed a novel method integrating uncertainty estimation to identify erroneous segmentation masks generated by CNNs.
  • Introduced a multi-task cross-task learning consistency approach to enforce correlation between pixel-level segmentation and geometric-level distance map tasks.
  • Conducted extensive experiments using varied labeled data quantities on the MICCAI 2017 ACDC Challenge Dataset for cardiac MRI segmentation.

Main Results:

  • Demonstrated the effectiveness of the proposed model for segmentation and uncertainty estimation of left ventricle (LV), right ventricle (RV), and myocardium (Myo).
  • Showcased the model's ability to evaluate the correlation between uncertainty and segmentation errors.
  • Validated the model's performance across different cardiac phases (end-diastole and end-systole) and with varying amounts of training data.

Conclusions:

  • The developed uncertainty estimation method effectively detects failures in DL-based segmentation masks for cardiac MRI.
  • The study provides a proof-of-concept for correlating uncertainty measures with segmentation errors, highlighting potential for flagging low-quality outputs.
  • This approach enhances the reliability of DL models in clinical settings by identifying potentially erroneous segmentation masks.