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

Respiration-coordinated attentional switch from feedforward to top-down informational flow directed by the basal forebrain: layer-specific blanket inhibition of pyramidal cells by neurogliaform cells in the piriform cortex.

Frontiers in neural circuits·2026
Same author

Balloon-occluded Alternative Infusion of Fragmented Gelatin Particles of TACE for Hepatocellular Carcinoma Refractory to Atezolizumab-Bevacizumab.

Anticancer research·2026
Same author

How do we think and what is the neural circuit mechanism for it? Possible roles of working memory and inner speech in thinking.

Frontiers in human neuroscience·2026
Same author

Deep learning-based artificial intelligence can improve the diagnosis of small bowel obstruction: stratified comparison study and hierarchical Bayesian model.

Scientific reports·2026
Same author

Endo-PairGS: pair priors for dynamic endoscopic scene reconstruction.

International journal of computer assisted radiology and surgery·2026
Same author

Automated Measurement of Occipito-Axial Angle on Cervical Radiographs Using a Deep Learning Object Detection Model: A Proof-of-Concept Study.

Neurospine·2026

Related Experiment Video

Updated: Dec 28, 2025

Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

603

Realistic endoscopic image generation method using virtual-to-real image-domain translation.

Masahiro Oda1, Kiyohito Tanaka2, Hirotsugu Takabatake3

  • 1Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8601, Japan.

Healthcare Technology Letters
|February 11, 2020
PubMed
Summary
This summary is machine-generated.

This study introduces a new method for generating realistic endoscopic images for simulation systems. Advanced deep learning models significantly improved the visual fidelity of virtual endoscopic simulations.

Keywords:
biological organscomputerised tomographydata visualisationdeep U-Netendoscope insertionsendoscopesendoscopic diagnosisendoscopic simulation systemsendoscopic treatmenthigh-quality image-domain translation resultsimage cleansingimage segmentationimage-domain translatormedical image processingnonrealistic virtual endoscopic imagesreal endoscopic imagesrealistic endoscopic image generation methodrealistic image generation methodrealistic imagesrendering (computer graphics)shallow U-Netunpaired virtual imagesvirtual reality

More Related Videos

Mixed Reality Assisted Radical Endoscopic Thyroidectomy
08:06

Mixed Reality Assisted Radical Endoscopic Thyroidectomy

Published on: January 31, 2025

602
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.5K

Related Experiment Videos

Last Updated: Dec 28, 2025

Photorealistic Learned Landscapes for Augmented Reality
06:54

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

603
Mixed Reality Assisted Radical Endoscopic Thyroidectomy
08:06

Mixed Reality Assisted Radical Endoscopic Thyroidectomy

Published on: January 31, 2025

602
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.5K

Area of Science:

  • Medical Simulation
  • Computer Vision
  • Medical Imaging

Background:

  • Endoscopic procedures are common in healthcare but carry risks.
  • Current endoscopic simulation systems lack realistic visuals, limiting their training effectiveness.
  • Realistic virtual endoscopic images are crucial for improving simulation-based training and reducing complications.

Purpose of the Study:

  • To develop a realistic image generation method for endoscopic simulation systems.
  • To enhance the visual fidelity of virtual endoscopic images generated from CT data.
  • To improve the overall value and effectiveness of endoscopic simulation training.

Main Methods:

  • Utilized volume rendering of patient CT scans to generate initial virtual endoscopic images.
  • Employed a virtual-to-real image-domain translation technique using a fully convolutional network (FCN).
  • Trained the FCN with unpaired virtual and real endoscopic images, minimizing cycle consistency loss.
  • Investigated various U-Net architectures (shallow, standard, deep, residual) as image-domain translators.
  • Preprocessed real endoscopic images through an image cleansing technique for higher quality training data.

Main Results:

  • The deep U-Net and U-Net with residual units demonstrated superior performance in generating realistic virtual endoscopic images.
  • The image-domain translation technique successfully enhanced the realism of virtual endoscopic visuals.
  • Image cleansing of real endoscopic datasets contributed to high-quality translation outcomes.

Conclusions:

  • The proposed method, particularly using deep U-Net architectures, significantly enhances the realism of virtual endoscopic images.
  • This advancement can lead to more effective training and rehearsal for endoscopic procedures.
  • Improved endoscopic simulation systems have the potential to reduce complications associated with real-world endoscope insertions.