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

Screening Mammography Completion Among Women Enrolled in a Lung Cancer Screening Program.

Journal of the American College of Radiology : JACR·2026
Same author

Comparison of simulated transient elastography sample volumes and global measurements on MRE and MRI-PDFF in patients with suspected MASLD.

Clinical imaging·2026
Same author

Effects of Pressure Cycling, Concentration, and Excitation Amplitude on Ambient Pressure Sensitivity of Perfluorobutane Microbubbles.

Ultrasound in medicine & biology·2026
Same author

Assessment of Early Breast Cancer Response to Chemotherapy with Ultrasound Radiomics.

Diagnostics (Basel, Switzerland)·2026
Same author

ACR Appropriateness Criteria® Breast Imaging During Lactation.

Journal of the American College of Radiology : JACR·2026
Same author

Controlled Antibiotic Release From Emulsion-Loaded Alginate and Fibrin Hydrogels Using Ultrasound.

Journal of biomedical materials research. Part A·2026

Related Experiment Video

Updated: Jun 22, 2025

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy
07:51

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy

Published on: June 1, 2019

19.4K

Characterizing Sentinel Lymph Node Status in Breast Cancer Patients Using a Deep-Learning Model Compared With

Priscilla Machado1, Aylin Tahmasebi1, Samuel Fallon2

  • 1Department of Radiology, Thomas Jefferson University, Philadelphia, PA.

Ultrasound Quarterly
|July 3, 2024
PubMed
Summary

A deep learning model demonstrated superior diagnostic performance in distinguishing benign from malignant sentinel lymph nodes (SLNs) in breast cancer patients compared to radiologists. The AI model showed improved accuracy, especially with balanced datasets, highlighting its potential in cancer diagnostics.

More Related Videos

Ultrasonographic Evaluation of Breast Cancer-related Lymphedema
05:44

Ultrasonographic Evaluation of Breast Cancer-related Lymphedema

Published on: January 12, 2017

10.0K
Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma
08:18

Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma

Published on: September 8, 2021

2.9K

Related Experiment Videos

Last Updated: Jun 22, 2025

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy
07:51

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy

Published on: June 1, 2019

19.4K
Ultrasonographic Evaluation of Breast Cancer-related Lymphedema
05:44

Ultrasonographic Evaluation of Breast Cancer-related Lymphedema

Published on: January 12, 2017

10.0K
Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma
08:18

Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma

Published on: September 8, 2021

2.9K

Area of Science:

  • Oncology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Accurate assessment of sentinel lymph nodes (SLNs) is crucial for breast cancer staging and treatment planning.
  • Current diagnostic methods for SLNs can be limited by inter-observer variability and subjective interpretation.
  • Deep learning offers a potential avenue for improving the objectivity and accuracy of SLN analysis.

Purpose of the Study:

  • To develop and evaluate a deep learning model for differentiating benign from malignant SLNs in breast cancer patients.
  • To compare the diagnostic performance of the AI model against the assessments of experienced radiologists.
  • To investigate the impact of dataset distribution on the AI model's performance.

Main Methods:

  • Seventy-nine breast cancer patients underwent lymphosonography and contrast-enhanced ultrasound (CEUS) to identify SLNs.
  • Google AutoML was utilized to create an image classification model using grayscale and CEUS images.
  • The dataset was divided into training (80%) and testing (20%) sets, with two models developed: one with all SLNs and another with a balanced distribution of benign and malignant SLNs.

Main Results:

  • The AI model achieved high diagnostic performance, with Area Under Precision-Recall Curve (AuPRC) values reaching 0.91 for CEUS images in Model 1 (all SLNs).
  • Statistically significant differences were observed between the AI model's performance and radiologists' assessments across different models and ultrasound modes (P < 0.001 for Model 1 CEUS AI vs. readers).
  • Inter-reader agreement among radiologists was low (κ = 0.20 for grayscale, 0.17 for CEUS), indicating variability in their assessments.

Conclusions:

  • The deep learning model, particularly Google AutoML, demonstrated enhanced diagnostic performance for SLN assessment, especially with balanced datasets.
  • The AI model's accuracy in differentiating benign and malignant SLNs shows promise for clinical application in breast cancer management.
  • Radiologist performance was not significantly influenced by the dataset's distribution, underscoring the AI's potential to provide consistent and objective analysis.