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

Effects of size and atomic defects on the buckling of single-layer and double-layer graphene nanoplates using AFEM.

Scientific reports·2026

Same author

Deep learning-based femoral reconstruction from intraoperative point clouds for enhanced knee arthroplasty registration.

International journal of computer assisted radiology and surgery·2026

Same author

Characterizing forearm skeletal muscle composition and function in breast cancer-related lymphedema using B-mode ultrasonography.

Clinical physiology and functional imaging·2026

Same author

Statistical shape model-based estimation of registration error in computer-assisted total knee arthroplasty.

International journal of computer assisted radiology and surgery·2026

Same author

From Speech to Sonography: Spectral Networks for Ultrasound Microstructure Classification.

IEEE transactions on bio-medical engineering·2025

Same author

Unraveling the thermal decomposition and chemical ionization of methionine using ion mobility spectrometry and computational chemistry.

Scientific reports·2025

Search research articles

Related Experiment Video

Updated: Dec 22, 2025

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

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

Published on: April 14, 2023

Two-stage ultrasound image segmentation using U-Net and test time augmentation.

Mina Amiri¹, Rupert Brooks^2,3, Bahareh Behboodi²

¹Concordia University, 1493 Saint-Catherine St W, Montreal, Quebec, Canada. amirim@encs.concordia.ca.

International Journal of Computer Assisted Radiology and Surgery

|May 1, 2020

Summary

This study introduces a two-stage approach for segmenting breast lesions in ultrasound images. By first detecting lesions and then segmenting them, the method significantly improves segmentation accuracy, especially for challenging cases.

Keywords:

Detection Segmentation U-Net Ultrasound

More Related Videos

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Related Experiment Videos

Last Updated: Dec 22, 2025

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

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

Published on: April 14, 2023

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Area of Science:

Medical Imaging
Computer-Aided Diagnosis
Artificial Intelligence

Background:

Breast lesion detection and segmentation in ultrasound images are crucial for computer-aided diagnosis.
Current automatic methods struggle with ultrasound artifacts and complex lesion characteristics, leaving segmentation an open problem.

Purpose of the Study:

To enhance the accuracy of breast lesion segmentation in ultrasound images.
To propose a two-stage approach incorporating a lesion detection stage prior to segmentation.

Main Methods:

Utilized a dataset of 163 breast ultrasound images containing benign lesions or malignant tumors.
Employed a U-Net architecture for initial lesion detection, followed by another U-Net for segmentation of the detected region.
Developed a test-time augmentation technique to refine the detection stage performance.

Main Results:

The two-stage approach improved the overall average Dice score by 1.8%.
For images with an initial Dice score below 70%, the average Dice score improved by 14.5%.
Accurate lesion detection was shown to substantially improve segmentation performance.

Conclusions:

The proposed two-stage technique demonstrates promising results for segmenting breast ultrasound images.
This method significantly reduces the likelihood of segmentation failure.