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

How is Bias Learned in Medical Image Analysis Models? An Exploration of the Encoding of Demographic Information in Deep Learning Models Trained to Detect Abnormalities on Chest X-Rays.

Journal of imaging informatics in medicine·2026
Same author

A trial of risk-adapted prostate cancer screening in a federally supported health center network serving a high-risk population.

Cancer·2026
Same author

Development of a Deep Learning Model to Automatically Identify Palatal Landmarks on Three-Dimensional Maxillary Dental Casts.

International journal of dentistry·2026
Same author

Cholesterol efflux protein, ABCA1, supports anticancer functions of myeloid immune cells.

Science advances·2026
Same author

Cholesterol efflux protein, ABCA1, supports anti-cancer functions of myeloid immune cells.

bioRxiv : the preprint server for biology·2025
Same author

Reduced cervical sampling of hysterectomy specimens with negative margins on conization: An opportunity to improve resource utilization.

American journal of clinical pathology·2024
Same journal

Comprehensive Profiling of Cell Surface Proteins in Testicular Germ Cell Tumors.

Cancer research communications·2026
Same journal

Tissue context shapes distinct premalignant outcomes in a HPV16 E6/E7-mutant Pik3ca transgenic mouse model.

Cancer research communications·2026
Same journal

Association of Cumulative Proton Pump Inhibitor Use with Prostate Cancer Risk and Outcomes: A Population-Based Cohort Study.

Cancer research communications·2026
Same journal

A phase 1b study of chemo-immunotherapy with pegylated liposomal doxorubicin and Pembrolizumab in estrogen receptor positive metastatic breast cancer.

Cancer research communications·2026
Same journal

DNA Methyltransferase Inhibition Prevents Platinum-Induced Ovarian Cancer Stem Cell Enrichment.

Cancer research communications·2026
Same journal

Chemotherapy Induced Neutropenia in Palestinian Patients with Solid Malignancies: Patient Characteristics, Severity Risk Factors & Management Patterns.

Cancer research communications·2026
See all related articles

Related Experiment Video

Updated: Jun 4, 2025

Optimization of a Multiplex RNA-based Expression Assay Using Breast Cancer Archival Material
11:12

Optimization of a Multiplex RNA-based Expression Assay Using Breast Cancer Archival Material

Published on: August 1, 2018

8.0K

Deep Learning Predicts Subtype Heterogeneity and Outcomes in Luminal A Breast Cancer Using Routinely Stained

Nikhil Cherian Kurian1,2, Peter H Gann3, Neeraj Kumar4

  • 1Department of Electrical Engineering, Indian Institute of Technology-Bombay, Mumbai, India.

Cancer Research Communications
|December 31, 2024
PubMed
Summary
This summary is machine-generated.

Deep learning quantifies breast cancer subtype admixture from routine images, revealing associations with aggressiveness and shorter progression-free survival. This scalable method aids research into intratumor heterogeneity (ITH) for precision oncology.

More Related Videos

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.7K
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
08:40

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

Published on: April 8, 2016

12.7K

Related Experiment Videos

Last Updated: Jun 4, 2025

Optimization of a Multiplex RNA-based Expression Assay Using Breast Cancer Archival Material
11:12

Optimization of a Multiplex RNA-based Expression Assay Using Breast Cancer Archival Material

Published on: August 1, 2018

8.0K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.7K
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
08:40

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

Published on: April 8, 2016

12.7K

Area of Science:

  • Pathology and Computational Biology
  • Genomics and Molecular Oncology
  • Artificial Intelligence in Medicine

Background:

  • Intratumor heterogeneity (ITH) complicates precision oncology treatments.
  • Existing methods for spatial quantification of ITH are not scalable for population-level studies.
  • Subtype admixture in breast cancer can be measured from transcriptomic data.

Purpose of the Study:

  • To develop and validate a deep learning model for quantifying subtype ITH in Luminal A (LumA) breast cancer from whole-slide images.
  • To investigate the association between image-detected subtype admixture and tumor aggressiveness, clinical characteristics, and patient outcomes.
  • To establish a scalable and cost-effective method for ITH research in breast cancer.

Main Methods:

  • Trained a deep convolutional neural network on routinely stained whole-slide images to quantify PAM50 subtype admixture in LumA breast cancer.
  • Utilized transcriptomic data and matrix factorization to determine subtype adherence in 680 TCGA-BRCA cases.
  • Validated the model's performance on a held-out set of 230 LumA-assigned cases.

Main Results:

  • Image-based subtype admixture correlated with increased HER2 positivity, tumor size, grade, and advanced tumor-node-metastasis stage.
  • Admixture was associated with altered mutation profiles (more TP53, fewer PIK3CA).
  • Higher levels of admixture significantly predicted shorter progression-free survival.

Conclusions:

  • Deep learning can accurately quantify subtype admixture in LumA breast cancer from standard histology images.
  • Image-detected ITH is clinically significant and associated with adverse outcomes.
  • This cost-effective, scalable approach offers a valuable tool for ITH research and potentially improving precision oncology.