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

Cross-Center Surgical Step Recognition in Standardized Training Tasks: Dataset, Baselines, and Transfer Analysis.

IEEE transactions on bio-medical engineering·2026
Same author

Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks.

Journal of imaging·2026
Same author

Perpendicular Vascular Changes in NBI-CE of Laryngeal Lesions: Diagnostic Accuracy, Reproducibility, and Common Pitfalls.

Diagnostics (Basel, Switzerland)·2025
Same author

Heart Rate and Heart Rate Variability as Indicators for Stress and Skill Development in Minimally-Invasive Surgery.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Analysis of intra- and inter-observer variability in 4D liver ultrasound landmark labeling.

Journal of medical imaging (Bellingham, Wash.)·2025
Same author

Real-time deformable structure tracking in 3D ultrasound sequences using deformable convolutional layers.

Computers in biology and medicine·2025

Related Experiment Video

Updated: Aug 25, 2025

Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver
04:33

Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver

Published on: August 21, 2018

10.5K

Landmark tracking in 4D ultrasound using generalized representation learning.

Daniel Wulff1, Jannis Hagenah2, Floris Ernst3

  • 1Institute for Robotics and Cognitive Systems, University of Lübeck, Ratzeburger Allee 160, Lübeck, 23562, Schleswig-Holstein, Germany. wulff@rob.uni-luebeck.de.

International Journal of Computer Assisted Radiology and Surgery
|October 15, 2022
PubMed
Summary

This study introduces a new 4D ultrasound target tracking method using a representation space and sliced-Wasserstein autoencoders. The approach achieves accurate tracking, even with anatomical deformations, offering improved generalizability across patients.

Keywords:
Greedy searchRadiotherapyRepresentation spaceSliced-Wasserstein autoencoder

More Related Videos

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
07:43

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy

Published on: July 2, 2021

3.1K
An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging
16:01

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging

Published on: September 24, 2017

10.5K

Related Experiment Videos

Last Updated: Aug 25, 2025

Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver
04:33

Three-Dimensional Ultrasonic Needle Tip Tracking with a Fiber-Optic Ultrasound Receiver

Published on: August 21, 2018

10.5K
In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
07:43

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy

Published on: July 2, 2021

3.1K
An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging
16:01

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging

Published on: September 24, 2017

10.5K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Biomedical Engineering

Background:

  • 4D ultrasound imaging is crucial for real-time anatomical visualization.
  • Traditional target tracking in 4D ultrasound relies on image patch similarity.
  • Deformations in anatomical structures pose challenges for accurate tracking.

Purpose of the Study:

  • To present and validate a novel target tracking concept for 4D ultrasound.
  • To replace traditional image patch similarity metrics with distances in a latent representation space.
  • To map 3D ultrasound patches into a representation space using sliced-Wasserstein autoencoders.

Main Methods:

  • A novel 4D ultrasound target tracking method operating in a representation space.
  • Unsupervised training of sliced-Wasserstein autoencoders to map 3D ultrasound patches.
  • A greedy algorithm approach utilizing distances between representation vectors for target relocation.
  • Validation on an in vivo dataset of liver images.
  • Exploration of three autoencoder training concepts for cross-patient generalizability.

Main Results:

  • Successful tracking in all eight annotated 4D ultrasound sequences.
  • A mean tracking error of 3.23 mm achieved with generalized fine-tuned autoencoders.
  • Demonstrated superior tracking performance of generalized autoencoders with fine-tuning compared to subject-individual autoencoders.

Conclusions:

  • Distances in a representation space effectively measure image patch similarity, even with anatomical deformations.
  • The proposed tracking algorithm is validated in an in vivo setting.
  • Generalized autoencoders, fine-tuned on minimal individual patient data, yield promising tracking results.