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

The role of digital health interventions for adults with tuberculosis: a network meta-analysis of randomized controlled trials.

Frontiers in medicine·2026

Same author

Osteopontin as an emerging multifunctional food ingredient: recent advancements, production, applications, and challenges.

Critical reviews in food science and nutrition·2026

Same author

Effectiveness of long-term graduated compression stockings after deep vein thrombosis: a GRADE-based meta-analysis of randomized controlled trials.

Frontiers in medicine·2026

Same author

Immune cell contributions to sarcopenia: Evidence from genetic analyses.

Medicine·2026

Same author

POSITION: Open World 3D Scene CAD Recomposition.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same author

High-clockrate free-space optical in-memory computing.

Light, science & applications·2026

Same journal

A computational model of chemically- and mechanically-induced thrombus formation in cerebral aneurysms.

Computers in biology and medicine·2026

Same journal

An improved catch fish optimization based deep learning model for Parkinson disease classification using EEG signal.

Computers in biology and medicine·2026

Same journal

Assessing the robustness of evaluation metrics for synthetic ECG signal quality.

Computers in biology and medicine·2026

Same journal

Integrating stemness and epithelial-mesenchymal transition signatures with machine learning identifies RUNX1 as a therapeutic vulnerability in colorectal cancer.

Computers in biology and medicine·2026

Same journal

Differential regional textural attributes of tongue in normal and acidity patients in the light of traditional Chinese medicine.

Computers in biology and medicine·2026

Same journal

SC-MSDNet: Spatial-consistent multi-view self-distillation for retinal OCT classification.

Computers in biology and medicine·2026

See all related articles

Search research articles

Related Experiment Video

Updated: May 14, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Optimized multiple instance learning for brain tumor classification using weakly supervised contrastive learning.

Kaoyan Lu¹, Shiyu Lin¹, Kaiwen Xue²

¹Key Laboratory of Atomic and Subatomic Structure and Quantum Control (Ministry of Education), Guangdong Basic Research Center of Excellence for Structure and Fundamental Interactions of Matter, School of Physics, South China Normal University, 378 Waihuan West Road, Panyu District, Guangzhou, 510006, Guangdong Province, China; Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials, Guangdong-Hong Kong Joint Laboratory of Quantum Matter, South China Normal University, 378 Waihuan West Road, Panyu District, 510006, Guangdong Province, Guangzhou, China.

Computers in Biology and Medicine

|April 12, 2025

Summary

This summary is machine-generated.

This study introduces a novel multi-instance learning (MIL) approach with weakly supervised contrastive learning for brain tumor classification from whole-slide images (WSIs). The method enhances feature representation and spatial relation modeling for improved diagnostic accuracy.

Keywords:

Brain tumor Classification Contrastive learning Cross-detection Multiple instance learning

More Related Videos

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

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

Published on: August 16, 2020

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Related Experiment Videos

Last Updated: May 14, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

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

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

Published on: August 16, 2020

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Area of Science:

Computational pathology
Artificial intelligence in oncology
Digital pathology

Background:

Accurate histopathological classification of brain tumors is vital for patient prognosis and quality of life.
Multi-instance learning (MIL) is standard for analyzing whole-slide images (WSIs), but faces challenges with data redundancy and spatial relationship modeling.
Existing MIL methods struggle with feature extractor representation capabilities.

Purpose of the Study:

To propose an advanced multi-instance learning framework incorporating weakly supervised contrastive learning for brain tumor classification.
To address limitations in current MIL methods, including input/feature redundancy, spatial relation modeling, and feature extractor performance.

Main Methods:

A novel framework combining a cross-detection MIL aggregator (CDMIL) and a pseudo-label-based contrastive learning model (PSCL).
CDMIL integrates an internal patch anchoring module (IPAM), local structural learning module (LSLM), and cross-detection module (CDM) for patch representation and fusion.
PSCL optimizes the feature encoder using pseudo-labels generated by IPAM, enhancing feature extraction for CDMIL.

Main Results:

The proposed method demonstrated superior performance compared to several state-of-the-art techniques on both a self-collected and a public dataset.
The framework effectively models spatial relationships between image patches and improves feature representation.
Bag-level contrastive loss was introduced to enhance interaction between different subtypes in the feature space.

Conclusions:

The developed framework offers a significant advancement in computational pathology for brain tumor classification.
The integration of CDMIL and PSCL provides a robust approach to overcome limitations of traditional MIL methods.
This approach holds promise for improving the accuracy and efficiency of brain tumor diagnosis through automated analysis of WSIs.