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

The Value of Five Novel Anthropometric Indicators in Evaluating MASLD and Liver Fibrosis in MASLD: A Cross-Sectional Study.

JGH open : an open access journal of gastroenterology and hepatology·2026
Same author

Efficacy and safety of multiple treatments for small hepatocellular carcinoma: an updated systematic review and component network meta analysis.

EClinicalMedicine·2026
Same author

Natural history and clinical characteristics of pulmonary subsolid nodules in pediatric patients: a cohort study.

Quantitative imaging in medicine and surgery·2026
Same author

Glymphatic system dysfunction in obstructive sleep apnea with mild cognitive impairment evidenced by DTI-ALPS.

Sleep medicine·2026
Same author

The micro-structural changes in white matter fibers associated with anxiety and depression in moderate-severe obstructive sleep apnea.

Scientific reports·2026
Same author

Association between lactate-to-albumin ratio and all-cause mortality in critically ill hepatic failure patients with sepsis: A retrospective cohort study.

Medicine·2026

Related Experiment Video

Updated: Aug 2, 2025

Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
10:25

Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning

Published on: April 12, 2024

1.5K

Focal Boundary Dice: Improved Breast Tumor Segmentation from MRI Scan.

Xiao-Xia Yin1, Yunxiang Jian1, Jing Shen2,3

  • 1Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou 510006, China.

Journal of Cancer
|April 14, 2023
PubMed
Summary

A new metric, Focal Boundary Dice, improves tumor segmentation by focusing on boundary accuracy and class imbalance. It outperforms standard measures, especially for smaller objects and boundary errors in MRI scans.

Keywords:
Intersection-over-Union loss, Tversky loss.Magnetic Resonance Imagingboundary binary cross-entropydeep learningdice lossmedical image segmentation

More Related Videos

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

22.6K
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K

Related Experiment Videos

Last Updated: Aug 2, 2025

Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
10:25

Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning

Published on: April 12, 2024

1.5K
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

22.6K
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K

Area of Science:

  • Medical Imaging Analysis
  • Computer Vision
  • Machine Learning

Background:

  • Accurate tumor segmentation in Magnetic Resonance Imaging (MRI) is crucial for diagnosis and treatment planning.
  • Existing evaluation metrics like Dice and Focal measures have limitations in assessing boundary quality and handling class imbalance.
  • Tumor segmentation models often struggle with accurately delineating tumor edges and differentiating between target and background areas.

Purpose of the Study:

  • To introduce Focal Boundary Dice, a novel segmentation evaluation measure designed to enhance the assessment of boundary quality and address class imbalance.
  • To update standard tumor segmentation evaluation protocols by proposing Focal Boundary Dice, improving upon existing methods.
  • To provide a more accurate and balanced evaluation metric suitable for various object sizes and error types in medical image segmentation.

Main Methods:

  • Developed Focal Boundary Dice by incorporating boundary quality and class imbalance considerations.
  • Conducted extensive analysis on MRI tumor segmentation data with varied object sizes and error types.
  • Introduced a boundary attention module to improve the extraction of tumor edge features.

Main Results:

  • Focal Boundary Dice demonstrated superior adaptiveness to boundary errors compared to standard Focal and Dice measures.
  • The new metric accurately evaluates segmentation performance across different scales without over-penalizing errors on smaller objects.
  • Experiments showed that Focal Boundary Dice identifies hard samples more accurately and offers balanced responsiveness across scales.

Conclusions:

  • Focal Boundary Dice offers a more suitable evaluation for segmentation tasks than classification-focused measures, improving accuracy in selecting hard samples.
  • The proposed metric facilitates boundary quality improvements and segmentation accuracy often overlooked by current Dice-based metrics and deep learning models.
  • Adoption of Focal Boundary Dice is expected to accelerate progress in segmentation methods and enhance classification accuracy in medical imaging.