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

X-ray Imaging01:24

X-ray Imaging

5.6K
German physicist Wilhelm Röntgen (1845–1923) was experimenting with electrical current when he discovered that a mysterious and invisible "ray" would pass through his flesh but leave an outline of his bones on a screen coated with a metal compound. In 1895, Röntgen made the first durable record of the internal parts of a living human: an "X-ray" image (as it came to be called) of his wife’s hand. Scientists worldwide quickly began their own experiments with...
5.6K
Imaging Studies for Cardiovascular System III: X-Ray01:20

Imaging Studies for Cardiovascular System III: X-Ray

193
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...
193

You might also read

Related Articles

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

Sort by
Same author

Evaluating AI-based mosquito monitoring technologies: a field study in Zhejiang Province, 2025.

Frontiers in veterinary science·2026
Same author

Monocular Markerless Motion Capture Enables Quantitative Assessment of Upper Extremity Reachable Workspace.

Sensors (Basel, Switzerland)·2026
Same author

Polygenic Stroke Risk and Autoimmune Disease: Associations and Clinical Modifiers.

IEEE transactions on bio-medical engineering·2026
Same author

Epidemiological and virological characteristics of influenza in Zhejiang Province, China: A 15-year surveillance study.

Human vaccines & immunotherapeutics·2026
Same author

Environmental, climatic, and social risk factors of severe fever with thrombocytopenia syndrome and the implications of climate change.

One health (Amsterdam, Netherlands)·2026
Same author

Cough Audio Recognition for Early Detection of Respiratory Diseases: Algorithm Development and Validation Study.

JMIR medical informatics·2026

Related Experiment Video

Updated: Jul 11, 2025

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
08:05

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

Published on: December 19, 2020

14.2K

Improving explainable AI with patch perturbation-based evaluation pipeline: a COVID-19 X-ray image analysis case

Jimin Sun1, Wenqi Shi2, Felipe O Giuste3

  • 1School of Computer Science and Engineering, Georgia Institute of Technology, Atlanta, 30322, USA.

Scientific Reports
|November 9, 2023
PubMed
Summary

This study introduces an automated patch perturbation method to evaluate explainable AI (XAI) in medical imaging. This approach removes the need for human experts, enabling better trust and selection of AI tools for clinical use.

More Related Videos

A Data-Driven Approach to Quantifying Immune States in Sepsis
07:42

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

179
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

2.8K

Related Experiment Videos

Last Updated: Jul 11, 2025

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
08:05

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

Published on: December 19, 2020

14.2K
A Data-Driven Approach to Quantifying Immune States in Sepsis
07:42

A Data-Driven Approach to Quantifying Immune States in Sepsis

Published on: February 7, 2025

179
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

2.8K

Area of Science:

  • Artificial Intelligence
  • Medical Imaging Analysis
  • Clinical Decision Support Systems

Background:

  • Explainable AI (XAI) methods are crucial for trust in clinical decision support systems, but their evaluation is hindered by a lack of automated strategies and human oversight.
  • Existing evaluation methods for XAI in medical imaging often require significant human intervention, additional annotations, and lack generalizability.

Purpose of the Study:

  • To propose an automated, patch perturbation-based approach for evaluating the quality of explanations generated by XAI methods in medical imaging analysis.
  • To eliminate the reliance on human experts and manual annotations in the XAI evaluation process.
  • To provide a comprehensive set of metrics for assessing XAI methods from multiple perspectives, including correctness, completeness, consistency, and complexity.

Main Methods:

  • Developed a patch perturbation approach using poisoning attacks with static and dynamic triggers during model retraining to automate XAI evaluation.
  • Introduced a comprehensive suite of evaluation metrics (correctness, completeness, consistency, complexity) applied during model inference.
  • Conducted a case study applying the proposed evaluation strategy to COVID-19 X-ray classification tasks using popular XAI methods.

Main Results:

  • The patch perturbation method successfully automates the evaluation of XAI quality without human intervention.
  • The proposed metrics provide a multi-faceted assessment of XAI methods, revealing their strengths and weaknesses.
  • The case study demonstrated the practical application and effectiveness of the automated evaluation strategy in a real-world medical imaging scenario.

Conclusions:

  • The proposed automated patch perturbation workflow offers a generalizable and efficient strategy for evaluating XAI methods in medical imaging.
  • This approach empowers developers to identify pitfalls and optimize XAI solutions, while assisting end-users in selecting suitable XAI tools for clinical practice.
  • Automated evaluation is essential for advancing the adoption and reliability of XAI in clinical decision support and translational research.