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

Electrospray Ionization Mass Spectrometry Can Provide Molecular-Level Insights into Key Clusters Related to Atmospheric New Particle Formation.

Analytical chemistry·2026

Same author

Population-scale characterization of the oral microbiome and associations with metabolic health.

Nature communications·2026

Same author

Leveraging Image-Text Pairs for Generalized Category Discovery in Medical Image Classification.

IEEE transactions on medical imaging·2026

Same author

mIT-CMCA: a cross-modal category alignment framework for robust maize disease identification.

Scientific reports·2026

Same author

Symmetry-Regulated Argentophilic Interactions via Homochiral Proline Dodecamers.

The journal of physical chemistry letters·2026

Same author

AI-based neoadjuvant immunotherapy response prediction across pan-cancer: a comprehensive review.

Cancer cell international·2025

Search research articles

Related Experiment Video

Updated: May 16, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Neighbor-Guided Unbiased Framework for Generalized Category Discovery in Medical Image Classification.

Wei Feng, Sijin Zhou, Yiwen Jiang

IEEE Journal of Biomedical and Health Informatics

|April 2, 2025

Summary

This study introduces a Neighbor-Guided Unbiased Framework (NGUF) to improve generalized category discovery (GCD) in medical imaging. NGUF reduces bias towards known categories, enhancing the identification of new disease classes from medical images.

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

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

Related Experiment Videos

Last Updated: May 16, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

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

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

Area of Science:

Artificial Intelligence
Medical Imaging Analysis
Machine Learning

Background:

Generalized category discovery (GCD) aims to identify novel semantic categories beyond those in the training data.
Existing GCD methods lack application in medical imaging for disease identification and suffer from prediction bias towards seen categories.
This bias leads to suboptimal clustering performance in medical diagnostic tasks.

Purpose of the Study:

To address the limitations of current GCD methods in medical image analysis.
To develop a framework that mitigates prediction bias and improves the discovery of new disease categories.
To enhance the accuracy and reliability of automated disease identification from medical images.

Main Methods:

Proposed a Neighbor-Guided Unbiased Framework (NGUF) for generalized category discovery in medical tasks.
Introduced a neighbor-guided cross-pseudo-clustering strategy leveraging nearest-neighbor information to adjust predictions and create unbiased supervision.
Implemented a view-invariant learning strategy for sample labeling and an adaptive weight learning strategy for dynamic prediction adjustment.
Incorporated a cross-batch knowledge distillation module to enhance training consistency.

Main Results:

NGUF effectively mitigates prediction bias in generalized category discovery for medical images.
The proposed framework demonstrates superior performance compared to state-of-the-art GCD algorithms on four medical image datasets.
Experimental results validate the efficacy of NGUF in improving the discovery of new disease categories.

Conclusions:

NGUF offers a novel and effective approach for unbiased generalized category discovery in medical imaging.
The method significantly improves the identification of previously unknown disease categories, aiding in disease understanding and diagnosis.
This framework represents a substantial advancement in applying GCD to critical medical diagnostic challenges.