Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Glucocorticoid Steroid and Alendronate Treatment Alleviates Dystrophic Phenotype with Enhanced Functional Glycosylation of α-Dystroglycan in Mouse Model of Limb-Girdle Muscular Dystrophy with FKRPP448L Mutation.

The American journal of pathology·2016

Same author

Exogenous H2S contributes to recovery of ischemic post-conditioning-induced cardioprotection by decrease of ROS level via down-regulation of NF-κB and JAK2-STAT3 pathways in the aging cardiomyocytes.

Cell & bioscience·2016

Same author

Common susceptibility variants are shared between schizophrenia and psoriasis in the Han Chinese population.

Journal of psychiatry & neuroscience : JPN·2016

Same author

Phylogenetic relationship of two popular edible Pleurotus in China, Bailinggu (P. eryngii var. tuoliensis) and Xingbaogu (P. eryngii), determined by ITS, RPB2 and EF1α sequences.

Molecular biology reports·2016

Same author

[Effect of ginsenoside total saponinon on regulation of P450 of livers of rats after γ-ray irradiation].

Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China journal of Chinese materia medica·2016

Same author

Lipid-polymer hybrid nanoparticles for the delivery of gemcitabine.

Journal of controlled release : official journal of the Controlled Release Society·2016

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

Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026

Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Aug 15, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

TrEnD: A transformer-based encoder-decoder model with adaptive patch embedding for mass segmentation in mammograms.

Dongdong Liu^1,2, Bo Wu^1,2, Changbo Li³

¹School of Biomedical Engineering, Capital Medical University, Beijing, China.

Medical Physics

|January 7, 2023

Summary

This summary is machine-generated.

This study introduces a deep learning model for automated breast mass segmentation in mammograms, achieving high accuracy. The TrEnD model significantly improves diagnostic assistance for breast cancer detection.

Keywords:

attention mechanism breast mass mammograms segmentation transformer

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

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

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

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Aug 15, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

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

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

Published on: November 30, 2022

Area of Science:

Medical Imaging
Artificial Intelligence
Oncology

Background:

Breast cancer is a leading cause of cancer in women, necessitating accurate mammogram analysis.
Radiologist interpretation of mammograms is time-intensive and error-prone, highlighting the need for automated solutions.
Computer-aided diagnostic (CAD) systems are crucial for efficient and reliable breast tumor detection and segmentation.

Purpose of the Study:

To develop a fully automatic and effective deep learning model for accurate breast mass segmentation in mammograms.
To address challenges in segmentation, including low contrast, varied shapes, and fuzzy boundaries of masses.
To enhance the performance of computer-aided diagnostic systems for breast cancer diagnosis.

Main Methods:

A transformer-based encoder-decoder model (TrEnD) was proposed, featuring adaptive patch embedding (APE) for optimized patch processing.
A hierarchical transformer-encoder and attention-gated-decoder structure was employed to refine feature extraction and suppress irrelevant background activations.
A dual-branch design was utilized for parallel extraction and fusion of global and local features, enhancing contextual understanding and information integrity.

Main Results:

The TrEnD model achieved high segmentation performance, with Dice coefficients of 92.20% and 91.83% on CBIS-DDSM and INbreast datasets, respectively.
Intersection over Union (IoU) scores reached 85.81% and 85.29% on the respective datasets, outperforming existing state-of-the-art methods.
The inclusion of APE and attention-gated modules led to significant improvements in Dice (6.54%) and IoU (10.07%) metrics.

Conclusions:

The proposed TrEnD network demonstrates effectiveness in automatic breast mass segmentation.
The model shows significant potential for assisting radiologists in clinical breast cancer diagnosis.
Quantitative and qualitative assessments confirm the robustness and efficacy of the developed segmentation approach.