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

Beyond auto-segmentation: the case for planning and dosimetry AI in head and neck radiation oncology.

BMJ oncology·2026
Same author

Early Salivary Gland Shrinkage Is Associated With an Increased Risk of Acute Xerostomia in Head and Neck Cancer Radiation Therapy.

Advances in radiation oncology·2026
Same author

LEARNABLE HIERARCHICAL VISUAL CONTEXTS FOR TUMOR SEGMENTATION IN COMPUTED TOMOGRAPHY IMAGES.

Proceedings. IEEE International Symposium on Biomedical Imaging·2026
Same author

Tumor-anchored deep feature random forests for out-of-distribution detection in lung cancer segmentation.

Transactions on machine learning research·2026
Same author

MHub.ai: A Standardized Platform for Reproducible AI Research in Medical Imaging.

Research square·2026
Same author

DUAL CROSS-ATTENTION SIAMESE TRANSFORMER FOR RECTAL TUMOR REGROWTH ASSESSMENT IN WATCH-AND-WAIT ENDOSCOPY.

Proceedings. IEEE International Symposium on Biomedical Imaging·2026
Same journal

A computer-aided diagnosis tool in prostate cancer patients with biochemical recurrence using 18F-PSMA PET/CT imaging.

Medical physics·2026
Same journal

Investigating the uncertainty of cellular microenvironment parameter estimations via diffusion MRI cytometry.

Medical physics·2026
Same journal

Correction to "On the shape of the radiation survival curve in tumor spheroids: The role of oxygen heterogeneity".

Medical physics·2026
Same journal

Multi-view constrained semi-supervised vertebra detection for 3D ultrasound spine volume.

Medical physics·2026
Same journal

Accuracy of quantitative <sup>177</sup>Lu SPECT/CT imaging: A systematic review.

Medical physics·2026
Same journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026
See all related articles

Related Experiment Video

Updated: Jan 3, 2026

Author Spotlight: Streamlined Brain and Skull Modeling for Enhanced Neurosurgical Planning in NHP Research
06:33

Author Spotlight: Streamlined Brain and Skull Modeling for Enhanced Neurosurgical Planning in NHP Research

Published on: February 9, 2024

1.7K

Patch-based generative adversarial neural network models for head and neck MR-only planning.

Peter Klages1, Ilyes Benslimane1, Sadegh Riyahi1

  • 1Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA.

Medical Physics
|November 17, 2019
PubMed
Summary
This summary is machine-generated.

Pix2pix and CycleGAN show promise for magnetic resonance (MR)-only radiotherapy planning in head and neck (HN) cancer. Combining overlapping synthetic CT (sCT) patch estimations improves accuracy, reducing errors for MR-only HN cancer treatment planning.

Keywords:
CycleGANMR-Guided Radiotherapyconditional generative adversarial networks (cGAN)generative adversarial networks (GAN)pix2pixsynthetic CT generation

More Related Videos

A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates
08:41

A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates

Published on: July 17, 2020

5.3K
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
08:25

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System

Published on: April 11, 2018

15.8K

Related Experiment Videos

Last Updated: Jan 3, 2026

Author Spotlight: Streamlined Brain and Skull Modeling for Enhanced Neurosurgical Planning in NHP Research
06:33

Author Spotlight: Streamlined Brain and Skull Modeling for Enhanced Neurosurgical Planning in NHP Research

Published on: February 9, 2024

1.7K
A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates
08:41

A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates

Published on: July 17, 2020

5.3K
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
08:25

Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System

Published on: April 11, 2018

15.8K

Area of Science:

  • Medical Imaging and Radiation Oncology
  • Artificial Intelligence in Healthcare
  • Deep Learning for Medical Image Synthesis

Background:

  • Magnetic resonance (MR) imaging is increasingly used for radiotherapy planning, but lacks electron density information crucial for dose calculation.
  • Synthetic computed tomography (sCT) generation from MR data is essential for MR-only treatment planning, especially in head and neck (HN) cancer.
  • Deep learning models like pix2pix and CycleGAN offer potential for accurate sCT synthesis.

Purpose of the Study:

  • To evaluate the accuracy of pix2pix and CycleGAN for patch-based synthetic CT (sCT) generation in head and neck (HN) cancer patients.
  • To assess the impact of various combination strategies for overlapping sCT patches on accuracy.
  • To determine the suitability of these methods for MR-only radiotherapy planning.

Main Methods:

  • Retrospective analysis of 23 MR/CT image pairs from HN cancer patients.
  • Training and evaluation of pix2pix and CycleGAN models for patch-based sCT generation.
  • Investigation of strategies including spatial context, data augmentation, and overlapping patch combination for HU estimation.

Main Results:

  • Pix2pix achieved lower mean absolute errors (MAE) than CycleGAN on both cross-validation and independent test sets (e.g., 94.0 HU vs 102.9 HU on independent set).
  • Combining overlapping sCT estimations reduced MAE and mean error (ME) compared to single-view methods.
  • Dosimetric accuracy was within 2% for planning target volumes (PTV) and critical structures; digitally reconstructed radiographs (DRRs) showed <1 mm agreement.

Conclusions:

  • Pix2pix and CycleGAN are promising for MR-only treatment planning in HN cancer, demonstrating good dosimetric accuracy and DRR similarity.
  • Combining overlapping patch estimations is an effective strategy to reduce sCT generation errors and potentially estimate transformation uncertainty.
  • Further studies with larger patient cohorts are needed due to the small sample size.