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

Search research articles

Related Experiment Video

Updated: Jan 12, 2026

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

Class-incremental learning using push-pull autoencoder for chest X-ray diagnosis.

Jayant Mahawar¹, Angshuman Paul¹

¹Department of Computer Science and Engineering, Indian Institute of Technology Jodhpur, N.H. 62, Nagaur Road, Karwar, Jodhpur, 342030, Rajasthan, India.

Computers in Biology and Medicine

|November 6, 2025

Summary

This summary is machine-generated.

Related Concept Videos

Imaging Studies for Cardiovascular System III: X-Ray

Imaging Studies for Cardiovascular System III: X-Ray

The most common cardiovascular diagnostic test is an X-ray. It produces images of the heart, blood vessels, and adjacent structures.
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...

You might also read

Related Articles

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

Sort by

Same author

Few-shot diagnosis of chest x-ray images using auxiliary information guided semi-deterministic infinite mixture prototypes.

Computers in biology and medicine·2025

Same author

LearnDiff: MRI image super-resolution using a diffusion model with learnable noise.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society·2025

Same author

Correcting Class Imbalances with Self-Training for Improved Universal Lesion Detection and Tagging.

ArXiv·2025

Same author

3D Universal Lesion Detection and Tagging in CT with Self-Training.

ArXiv·2025

Same author

Generalizable diagnosis of chest radiographs through attention-guided decomposition of images utilizing self-consistency loss.

Computers in biology and medicine·2024

Same author

An extremely lightweight CNN model for the diagnosis of chest radiographs in resource-constrained environments.

Medical physics·2023

Same journal

Rapid personalisation of cardiovascular models using invasively measured right ventricular pressure.

Computers in biology and medicine·2026

Same journal

Biologically inspired mechanisms for enhancing robustness in EEG signal modeling: Challenges, opportunities, and perspectives.

Computers in biology and medicine·2026

Same journal

Machine learning-based detection of missed inspiratory efforts using esophageal pressure during noisy pressure support ventilation.

Computers in biology and medicine·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

See all related articles

Class-incremental learning (CIL) models for chest X-ray diagnosis were improved using the Push-Pull Autoencoder (PPAE). PPAE reduces catastrophic forgetting, enhancing diagnostic accuracy for new and old diseases.

Area of Science:

Medical Imaging
Artificial Intelligence
Machine Learning

Background:

Class-incremental learning (CIL) enables models to learn new classes without prior data, crucial for evolving medical datasets.
Existing CIL models struggle with chest X-ray diagnosis, while deep learning models for X-rays often exhibit catastrophic forgetting in CIL settings.

Purpose of the Study:

To develop a novel CIL framework tailored for chest X-ray analysis, addressing the limitations of current models.
To mitigate catastrophic forgetting and improve diagnostic performance in incremental learning scenarios for medical imaging.

Main Methods:

Proposed the Push-Pull Autoencoder (PPAE), a model utilizing a dual latent space to disentangle abnormality-specific and abnormality-agnostic features.
Implemented a coreset generation algorithm to select representative exemplars, preserving knowledge from previous classes without full retraining.

Keywords:

Chest X-ray diagnosis Class-incremental learning Contrastive learning Coreset Push-pull autoencoder

Related Experiment Videos

Last Updated: Jan 12, 2026

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

Trained PPAE to refine feature representations by clustering similar samples and distinguishing dissimilar ones based on learned features.

Main Results:

Achieved up to 3% improvement in F1 score and 4% improvement in AUROC across diverse chest X-ray datasets.
Demonstrated significant reduction in catastrophic forgetting, maintaining high diagnostic accuracy on previously learned classes.
Validated the framework's robustness and effectiveness in continuous learning tasks for chest X-ray diagnosis.

Conclusions:

The PPAE framework effectively addresses catastrophic forgetting in class-incremental learning for chest X-ray diagnosis.
The proposed method enhances continuous learning capabilities, offering a promising approach for evolving medical diagnostic systems.
PPAE shows potential for advancing long-term, adaptive diagnostic solutions in medical imaging.