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

A nomogram for risk prediction in patients with heart failure and diabetes: Development and validation.

Medicine·2026
Same author

Extracellular Matrix-Related Gene Signature Implied Immunosuppressive Status and Adverse Survival in Diffuse Glioma Patients.

Cancers·2026
Same author

Enhanced Prediction of Cardiovascular Disease Through Integrated Machine Learning Models Combining Clinical and Demographic Characteristics.

Diagnostics (Basel, Switzerland)·2026
Same author

Molecular alterations prediction in gliomas via an interpretable deep learning model: a multicentre and retrospective study.

The Lancet. Digital health·2026
Same author

Unsupervised single-domain generalization for tissue classification via progressive domain transformation.

Medical image analysis·2026
Same author

Application of lauroyl arginate ethyl to control Listeria monocytogenes biofilms and cross-contamination via domestic kitchen surfaces.

Food research international (Ottawa, Ont.)·2026

Related Experiment Video

Updated: Jun 28, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K

Imaging segmentation mechanism for rectal tumors using improved U-Net.

Kenan Zhang1,2, Xiaotang Yang3, Yanfen Cui4

  • 1College of Electronic Information and Optical Engineering, Taiyuan University of Technology, Taiyuan, 030024, China.

BMC Medical Imaging
|April 23, 2024
PubMed
Summary

This study introduces an improved U-Net model for segmenting rectal tumors in MRI scans. The new method enhances accuracy by incorporating attention mechanisms, outperforming existing techniques for better cancer treatment planning.

Keywords:
MR imageRectal cancerSemantic segmentationU-Net

More Related Videos

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

393

Related Experiment Videos

Last Updated: Jun 28, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

393

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Accurate segmentation of cancerous regions in Magnetic Resonance Images (MRI) is crucial for effective radiation therapy.
  • Rectal cancer segmentation in MRI presents challenges due to surrounding organs with similar shapes and insufficient high-resolution features in conventional deep learning models.

Purpose of the Study:

  • To address the limitations of existing deep learning methods for rectal tumor segmentation in MRI.
  • To propose an improved U-Net segmentation network incorporating attention mechanisms to enhance segmentation accuracy.

Main Methods:

  • The study adapted the traditional U-Net architecture, integrating a ResNeSt module for feature extraction and a shape module post-encoder.
  • Attention mechanisms were employed to refine the network's learning process and improve segmentation precision.
  • The combined outputs from the shape module and decoder were utilized to generate segmentation results.

Main Results:

  • The proposed method was validated on a dataset of 3773 2D MRI scans from 304 patients.
  • Achieved superior performance with Dice, Mean Paired Accuracy (MPA), Mean Intersection over Union (MioU), and Frequency Weighted Intersection over Union (FWIoU) scores of 0.987, 0.946, 0.897, and 0.899, respectively.
  • Demonstrated statistically significant improvements compared to existing segmentation methods.

Conclusions:

  • The developed U-Net model with attention mechanisms offers an effective solution for rectal tumor segmentation in MRI.
  • The method saves time, allowing radiologists to focus on complex cases and ensuring high diagnostic quality and accuracy in cancer treatment planning.