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

A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates08:41

A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates

5.3K
The method outlined below aims to provide a comprehensive protocol for the preparation of nonhuman primate (NHP) neurosurgery using a novel combination of three-dimensional (3D) printing methods and MRI data...
5.3K
Constructing and Visualizing Models using Mime-based Machine-learning Framework06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

2.3K
Mime is a flexible computational framework to construct a machine learning-based integration model with elegant performance. Here, we provide a detailed step-by-step procedure for developing predictive models with high accuracy, leveraging complex datasets to identify critical genes associated with disease progression, patient outcomes, and therapeutic response.
2.3K
Long-patch Base Excision Repair01:02

Long-patch Base Excision Repair

7.9K
Since the discovery of the two BER pathways, there has been a debate about how a cell chooses one pathway over the other and the factors determining this selection. Numerous in vitro experiments have pointed out multiple determinants for the sub-pathway selection. These are:
7.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

10.6K
This is a method for training a multi-slice U-Net for multi-class segmentation of cryo-electron tomograms using a portion of one tomogram as a training input. We describe how to infer this network to other tomograms and how to extract segmentations for further analyses, such as subtomogram averaging and filament...
10.6K
Radiation Planning Assistant - A Web-based Tool to Support High-quality Radiotherapy in Clinics with Limited Resources05:18

Radiation Planning Assistant - A Web-based Tool to Support High-quality Radiotherapy in Clinics with Limited Resources

1.8K
This protocol describes a series of automated tools designed for high-quality radiotherapy autocontouring and autoplanning that are being packaged into a web-based service to maximize robustness and scalability while minimizing operational...
1.8K
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

1.5K
We present a protocol that combines recombinase polymerase amplification with a CRISPR/Cas12a system for trace detection of DNA viruses and builds portable smartphone microscopy with an artificial intelligence-assisted classification for point-of-care DNA virus...
1.5K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

PANTHER Challenge Report: Cross-Domain Pancreatic Tumor Segmentation in Magnetic Resonance Imaging.

Medical image analysis·2026
Same author

Correction: Making scientific publishing more balanced and reversing the current dangerous drift to better support global environmental conservation and planetary health.

Research integrity and peer review·2026
Same author

Is There Value in Performing Yearly Screening for Latent Tuberculosis Infection by Interferon-Gamma Release Assay Among Patients Living With HIV in Non-Endemic Settings?

Open forum infectious diseases·2026
Same author

Magnetic resonance imaging-guided radiotherapy for prostate cancer: A systematic review of the literature.

Cancer radiotherapie : journal de la Societe francaise de radiotherapie oncologique·2026
Same author

A new automated 3d facial soft tissue landmarking method via deep learning.

Journal of dentistry·2026
Same author

Standardizing MRI-only radiotherapy commissioning: Benchmark dataset and acceptance levels from the MESCAL initiative.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2026

Related Experiment Video

Updated: Jan 19, 2026

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

Comparison of Deep Learning-Based and Patch-Based Methods for Pseudo-CT Generation in MRI-Based Prostate Dose

Axel Largent1, Anaïs Barateau1, Jean-Claude Nunes1

  • 1Univ Rennes, CLCC Eugène Marquis, INSERM, LTSI - UMR 1099, F-35000 Rennes, France.

International Journal of Radiation Oncology, Biology, Physics
|September 11, 2019
PubMed
Summary

Deep learning methods (DLMs) generate pseudo-CT (pCT) for MRI-based radiotherapy. GAN L2 and U-Net L2 methods offer the lowest dose uncertainties and fast computation times for prostate cancer treatment planning.

More Related Videos

Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

2.3K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.6K

Related Experiment Videos

Last Updated: Jan 19, 2026

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
Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

2.3K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.6K

Area of Science:

  • Medical Imaging
  • Radiotherapy Physics
  • Artificial Intelligence in Medicine

Background:

  • Magnetic resonance imaging (MRI) based dose planning requires computed tomography (CT) data for electron density information.
  • Generating pseudo-CT (pCT) from MRI data using deep learning methods (DLMs) is an emerging technique to overcome CT limitations.
  • Evaluating the accuracy and efficiency of different DLMs for pCT generation is crucial for clinical implementation.

Purpose of the Study:

  • To evaluate and compare deep learning methods, specifically U-Net and generative adversarial network (GAN), for pseudo-CT (pCT) generation from MRI data.
  • To assess the impact of various loss functions (L2, perceptual loss) and a patch-based method (PBM) on pCT quality and dose calculation accuracy.
  • To determine the most effective DLM configuration for MRI-based dose planning in prostate cancer patients.

Main Methods:

  • Generated pCTs from T2-weighted MRIs of 39 prostate cancer patients using 4 GAN configurations (L2, single-scale PL, multiscale PL, weighted multiscale PL), 2 U-Net configurations (L2, single-scale PL), and a patch-based method (PBM).
  • Assessed imaging accuracy using mean absolute error and mean error in Hounsfield units compared to reference CT (CTref).
  • Quantified dose uncertainties by comparing dose volume histograms (DVHs) and analyzing 3D gamma index between CTref and pCT derived doses.

Main Results:

  • GAN L2 and U-Net L2 demonstrated the lowest mean absolute errors (≤34.4 HU) and minimal differences in DVH points compared to CTref.
  • Dose uncertainties for GAN L2 and U-Net L2 were low (≤0.6% for PTV V95%, ≤0.5% for rectum V70Gy, ≤0.1% for bladder V50Gy).
  • DLMs achieved significantly faster pCT generation times (15s) compared to PBM (62 min), with >99% gamma pass rates for all DLMs.

Conclusions:

  • Deep learning methods, particularly GAN L2 and U-Net L2, effectively generate pCTs for MRI-based dose planning with low image and dose uncertainties.
  • These DLMs offer a significant advantage in computation time compared to traditional methods, making them suitable for clinical workflows.
  • The study highlights the potential of DLMs to improve the accuracy and efficiency of radiotherapy planning using MRI-only data.