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

Oliceridine compared with sufentanil in multimodal analgesia after renal transplantation: protocol for a single-centre, randomised, double-blind, positive-controlled trial.

BMJ open·2026
Same author

Factors affecting the gut microbiota of rural residents under different physical activity levels: a cross-sectional study.

Frontiers in microbiology·2026
Same author

Association between intrinsic capacity, genetic predisposition, and risk of venous thromboembolism.

Journal of vascular surgery. Venous and lymphatic disorders·2026
Same author

Frontal Polymerization-Enabled Rapid Fabrication of Gelatin-Containing Hydrogels with Good Mechanical and Biological Properties.

Gels (Basel, Switzerland)·2026
Same author

Exercise modulates food reward: neurobiological mechanisms and implications for weight management.

Frontiers in nutrition·2026
Same author

Comparison of oxycodone hydrochloride injection versus sufentanil for postoperative analgesia in patients undergoing minimally invasive cardiac valve replacement: study protocol for a single-centre, randomised, double-blind non-inferiority trial.

BMJ open·2026

Related Experiment Video

Updated: Sep 19, 2025

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

6.9K

Deep learning for fine-grained molecular-based colorectal cancer classification.

Junyu Bian1, Yansong Li2, Yamei Dang3

  • 1The First School of Clinical Medicine, Lanzhou University, Lanzhou, China.

Translational Cancer Research
|June 18, 2025
PubMed
Summary

This study introduces a deep learning method using H&E images for colorectal cancer (CRC) molecular classification. The hybrid CNN-ViT model shows potential for faster, cost-effective CRC molecular detection.

Keywords:
Artificial intelligence (AI)colorectal cancer (CRC)deep learning (DL)whole slide image (WSI)

More Related Videos

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

201
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

3.0K

Related Experiment Videos

Last Updated: Sep 19, 2025

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

6.9K
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

201
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

3.0K

Area of Science:

  • Oncology
  • Computational Pathology
  • Bioinformatics

Background:

  • Colorectal cancer (CRC) is a leading cause of cancer deaths globally.
  • Molecular markers like MSI, BRAF, KRAS, and NRAS are crucial for CRC diagnosis.
  • Traditional molecular detection methods are expensive and time-consuming.

Purpose of the Study:

  • To develop a fine-grained classification method for CRC using hematoxylin and eosin (H&E) stained tissue images.
  • To leverage deep learning (DL) technology for novel insights into CRC molecular diagnosis.
  • To provide a potentially faster and more cost-effective alternative to traditional methods.

Main Methods:

  • A dataset of 383 CRC patient H&E images (LZUFH_CRC) was created.
  • A hybrid DL model combining Convolutional Neural Network (CNN) and Vision Transformer (ViT) was proposed.
  • A two-stage training strategy was employed, followed by performance evaluation and comparison.

Main Results:

  • The hybrid DL model achieved an accuracy (ACC) of 0.524 and AUC of 0.791 on the LZUFH_CRC dataset.
  • The model demonstrated strong performance for MSI (F1-score 0.724) and NRAS (F1-score 0.514) classification.
  • Feature activation maps highlighted the model's focus on tumor-stroma boundaries and mesenchymal areas, with no significant bias across clinical characteristics.

Conclusions:

  • A novel DL-based fine-grained classification method for CRC using H&E images was developed.
  • The study indicates DL technology holds promise for molecular detection in CRC.
  • Future work will focus on optimizing the model to enhance accuracy and efficiency for CRC classification.