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 Videos

Mutual information in coupled multi-shape model for medical image segmentation.

A Tsai1, W Wells, C Tempany

  • 1Laboratory for Information and Decision Systems (LIDS), Massachusetts Institute of Technology (MIT), Room #35-427, Cambridge, MA 02139, USA. atsai@mit.edu

Medical Image Analysis
|November 30, 2004
PubMed
Summary

This study introduces a novel multi-shape active contour model for simultaneous medical image segmentation. The advanced framework effectively segments multiple structures, improving accuracy and robustness in complex imaging data.

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

Quality of surgery in oncological trials: the patient's perspective.

European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology·2025
Same author

Rapid quasi-periodic oscillations in the relativistic jet of BL Lacertae.

Nature·2022
Same author

Human immune system adaptations to simulated microgravity revealed by single-cell mass cytometry.

Scientific reports·2021
Same author

Microscopic Imaging of the Stress Tensor in Diamond Using in Situ Quantum Sensors.

Nano letters·2019
Same author

MR imaging-guided prostate interventional imaging: Ready for a clinical use?

Diagnostic and interventional imaging·2018
Same author

Collaborating to Compete: Blood Profiling Atlas in Cancer (BloodPAC) Consortium.

Clinical pharmacology and therapeutics·2017

Area of Science:

  • Medical Image Analysis
  • Computer Vision
  • Computational Anatomy

Background:

  • Deformable active contour models are crucial for medical image segmentation.
  • Previous models struggled with simultaneous segmentation of multiple, potentially interacting shapes.
  • Accurate segmentation is vital for applications like image-guided radiotherapy.

Purpose of the Study:

  • To extend shape-based deformable active contour models for simultaneous multi-shape segmentation.
  • To develop a unified framework that captures inter-shape co-variations.
  • To enhance robustness against noise in medical imaging.

Main Methods:

  • Employed multiple signed distance functions for implicit representation of multiple shape classes.
  • Derived a parametric multi-shape model using principal component analysis on signed distance functions.

Related Experiment Videos

  • Utilized a single mutual information-based cost criterion to couple shape deformations.
  • Applied the framework to segment prostate, rectum, and internal obturator muscles in MR images.

    • Main Results:

    • The proposed model successfully segments multiple shapes simultaneously and seamlessly.
    • Effective utilization of co-dependencies among shapes improved segmentation accuracy.
    • The parametric model captured significant shape variability.
    • The algorithm demonstrated robustness to substantial additive noise.

    Conclusions:

    • The developed multi-shape active contour model offers a powerful tool for complex medical image segmentation.
    • Simultaneous segmentation leveraging inter-shape dependencies enhances performance and robustness.
    • This approach shows significant potential for improving image-guided interventions, such as prostate brachytherapy.