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

Light Acquisition02:16

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

An Inverse Signorini Obstacle Problem.

Archive for rational mechanics and analysis·2026
Same author

Intermittent Fasting Attenuates Cognitive Decline in D-Galactose-Induced Aging Rats in Association with β-Hydroxybutyrate and PI3K/AKT/GSK-3β Signaling.

Neurochemical research·2026
Same author

Deep learning model for pathological invasiveness prediction using smartphone-based surgical resection images in clinical stage IA lung adenocarcinoma (SuRImage): a prospective, multicentric, diagnostic study.

The Lancet. Digital health·2026
Same author

Evaluation of Biocontrol Efficacy of <i>Bacillus velezensis</i> HAB-2 Combined with <i>Pseudomonas hunanensis</i> and <i>Enterobacter soli</i> Against Cowpea Fusarium Wilt.

Microorganisms·2025
Same author

Characterization of <i>Ganoderma pseudoferreum</i> mitogenome revealed a remarkable evolution in genome size and composition of protein-coding genes.

Frontiers in plant science·2025
Same author

Uncertainty-Aware Survival Analysis With Dirichlet Distribution for Multi-Scale Pathology and Genomics.

IEEE transactions on medical imaging·2025

Related Experiment Video

Updated: Jul 6, 2026

Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

AttriMIL: Revisiting attention-based multiple instance learning for whole-slide pathological image classification

Linghan Cai1, Shenjin Huang2, Ye Zhang1

  • 1School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, 518055, China.

Medical Image Analysis
|May 17, 2025
PubMed
Summary

AttriMIL enhances whole-slide pathological image analysis by introducing attribute-aware multiple instance learning (MIL). This framework improves disease classification and region localization by better differentiating tissue instances.

Keywords:
Attribute scoring mechanismMultiple instance learningPathological image analysisPathology adaptive learningPathology attribute constraint

More Related Videos

Methods to Test Visual Attention Online
09:44

Methods to Test Visual Attention Online

Published on: February 19, 2015

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Jul 6, 2026

Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

Methods to Test Visual Attention Online
09:44

Methods to Test Visual Attention Online

Published on: February 19, 2015

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

Area of Science:

  • Computational pathology
  • Artificial intelligence in medicine
  • Digital pathology

Background:

  • Multiple instance learning (MIL) is crucial for whole-slide pathological image (WSI) analysis, especially with slide-level labels.
  • Attention-based MIL models advance weakly supervised WSI classification but struggle with instance differentiation, potentially degrading performance.
  • Differentiating between instances is key for accurate tissue identification and classification in WSI analysis.

Purpose of the Study:

  • To introduce AttriMIL, an attribute-aware multiple instance learning framework to address limitations in current MIL approaches for WSI analysis.
  • To enhance the differentiation of instances within gigapixel-resolution pathological images.
  • To improve the accuracy and robustness of weakly supervised WSI classification and disease-positive region localization.

Main Methods:

  • Developed a multi-branch attribute scoring mechanism to quantify pathological attributes of individual instances.
  • Introduced region-wise and slide-wise attribute constraints to model instance correlations dynamically during training.
  • Implemented a pathology adaptive learning technique to optimize pre-trained feature extractors for task-specific feature extraction.

Main Results:

  • AttriMIL consistently outperformed state-of-the-art methods across five public datasets.
  • Demonstrated superior performance in bag classification accuracy, generalization ability, and disease-positive region localization.
  • The attribute constraints effectively encouraged the network to capture spatial patterns and semantic similarities, improving sensitivity to challenging instances.

Conclusions:

  • AttriMIL provides a robust framework for attribute-aware multiple instance learning in WSI analysis.
  • The proposed attribute constraints and adaptive learning technique significantly enhance classification and localization performance.
  • AttriMIL offers a promising advancement for computational pathology, aiding clinical diagnosis through improved WSI analysis.