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

Survival Tree

Survival Tree

Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...

You might also read

Related Articles

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

Sort by

Same author

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification.

ArXiv·2026

Same author

<i>Radiology: Cardiothoracic Imaging</i> Highlights 2025.

Radiology. Cardiothoracic imaging·2026

Same author

A multi-agent large language model framework to automatically assess performance of a clinical AI Triage tool.

npj health systems·2026

Same author

OVERVIEW OF THE CXR-LT 2026 CHALLENGE: MULTI-CENTER LONG-TAILED AND ZERO SHOT CHEST X-RAY CLASSIFICATION.

Proceedings. IEEE International Symposium on Biomedical Imaging·2026

Same author

Rethinking radiology AI benchmarks.

Radiology advances·2026

Same author

Towards generalizable AI in medicine via Generalist-Specialist Collaboration.

Nature biomedical engineering·2026

Same journal

MetaboNet-Bench: A Multi-modal Benchmark for Glucose Forecasting in Type 1 Diabetes.

ArXiv·2026

Same journal

A Positron Range Correction with Texture Preservation Framework in PET Imaging.

ArXiv·2026

Same journal

Automated optimization of force field parameters against ensemble-averaged measurements with Bayesian Inference of Conformational Populations.

ArXiv·2026

Same journal

Droplet Fusion as a Relaxation Process: Comparison with Shape Recovery of Newtonian and Viscoelastic Droplets.

ArXiv·2026

Same journal

Ridge-filter crosstalk in conformal proton FLASH planning: dependence on beamlet pitch and iterative mitigation.

ArXiv·2026

Same journal

Electrochemical DNA Hairpin Sensors for Differentiating Small Molecule Intercalation from Minor Groove Binding.

ArXiv·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 15, 2025

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

How Does Pruning Impact Long-Tailed Multi-Label Medical Image Classifiers?

Gregory Holste¹, Ziyu Jiang², Ajay Jaiswal¹

¹The University of Texas at Austin, Austin, TX, USA.

|October 4, 2023

Summary

This summary is machine-generated.

Neural network pruning for medical image diagnosis can alter model behavior, especially for rare diseases. Radiologists found pruned model disagreements (PIEs) harder to diagnose, suggesting careful deployment is needed.

Keywords:

Chest X-Ray Imbalance Long-Tailed Learning Pruning

More Related Videos

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Related Experiment Videos

Last Updated: Jul 15, 2025

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

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Area of Science:

Artificial Intelligence
Medical Imaging
Computer Vision

Background:

Deep neural network pruning reduces model size and inference time.
Pruning's impact on models for long-tailed, multi-label clinical data is poorly understood.
This gap poses risks for diagnostic model deployment in healthcare.

Approach:

Analyzed pruning effects on neural networks diagnosing thorax diseases from chest X-rays (CXRs).
Examined disease-specific impacts and class 'forgettability' based on frequency and co-occurrence.
Identified pruning-identified exemplars (PIEs) where models disagree and conducted a human reader study.

Key Points:

Pruning affects different diseases unevenly, with rarer diseases being more susceptible.
Radiologists perceived PIEs as having more label noise, lower image quality, and higher diagnostic difficulty.
Class forgettability correlates with disease frequency and co-occurrence patterns.

Conclusions:

This study is the first to analyze pruning's impact on neural networks for multi-label CXR diagnosis.
Findings highlight the need for careful consideration of pruning in clinical AI deployment.
Understanding pruning's effects is crucial for safe and reliable medical AI.