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

Exploring key genes and prognostic validation of prostate cancer through transcriptome-wide association study.

Scientific reports·2026

Same author

A knowledge-guided deep learning framework for quantitative nucleic acid testing.

Briefings in bioinformatics·2026

Same author

A review of deep learning approaches for drug synergy prediction in cancer.

npj drug discovery·2026

Same author

Interfacial Morphology Shapes Endothelial Organization and Function in a Blood-Brain Barrier-on-a-Chip.

ACS applied materials & interfaces·2026

Same author

Cervical Anastomotic Leakage After Minimally Invasive McKeown Esophagectomy in the SAFER Surgical System: Clinical Course and Predictors of Delayed Healing.

Thoracic cancer·2026

Same author

Pathogenicity prediction for noncanonical splice-altering variants based on multimodal feature fusion.

Briefings in bioinformatics·2026

Search research articles

Related Experiment Video

Updated: Aug 28, 2025

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

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

Published on: April 8, 2016

Divide-and-Attention Network for HE-Stained Pathological Image Classification.

Rui Yan^1,2, Zhidong Yang¹, Jintao Li¹

¹High Performance Computer Research Center, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100045, China.

|September 14, 2022

Summary

This summary is machine-generated.

A novel Divide-and-Attention Network (DANet) improves pathological image classification by decomposing images into nuclei and non-nuclei components. This deep learning approach adaptively focuses on key features, enhancing diagnostic accuracy for cancer grading.

Keywords:

attention mechanism convolutional neural network knowledge embedding pathological image classification

More Related Videos

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

Automated Dissection Protocol for Tumor Enrichment in Low Tumor Content Tissues

Automated Dissection Protocol for Tumor Enrichment in Low Tumor Content Tissues

Published on: March 29, 2021

Related Experiment Videos

Last Updated: Aug 28, 2025

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

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

Published on: April 8, 2016

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

Automated Dissection Protocol for Tumor Enrichment in Low Tumor Content Tissues

Automated Dissection Protocol for Tumor Enrichment in Low Tumor Content Tissues

Published on: March 29, 2021

Area of Science:

Digital Pathology
Computational Pathology
Medical Image Analysis

Background:

Convolutional Neural Networks (CNNs) struggle with fine-grained classification of pathological images due to differences from natural images.
Clinical practice often involves decomposing pathological images for better diagnosis.

Purpose of the Study:

To propose a novel deep learning model, the Divide-and-Attention Network (DANet), for improved classification of Hematoxylin-and-Eosin (HE)-stained pathological images.
To leverage image decomposition and attention mechanisms to enhance feature learning for pathological image analysis.

Main Methods:

The DANet decomposes pathological images into nuclei and non-nuclei components using deep learning.
Independent feature learning is performed on each component, followed by adaptive focus on important features via branch selection attention.
Deep Canonical Correlation Analysis (DCCA) constraints are introduced for branch fusion attention, maximizing inter-branch correlation.

Main Results:

The DANet achieved superior performance on three datasets.
Average classification accuracy reached 92.5% for breast cancer classification.
Accuracies of 95.33% for colorectal cancer grading and 91.6% for breast cancer grading were obtained.

Conclusions:

The proposed DANet effectively improves pathological image classification by decomposing images and adaptively focusing on salient features.
The integration of DCCA constraints further enhances feature fusion, emphasizing specific tissue structures for more accurate diagnoses.