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

You might also read

Related Articles

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

Sort by
Same author

Global research trends, knowledge structure, and future directions in melioidosis research: a bibliometric analysis, 1990-2025.

Tropical medicine and health·2026
Same author

How does digital financial inclusion affect the health of older adults? Evidence from China.

SSM - population health·2026
Same author

Desloratadine Rescues Schizophrenia-like Phenotypes by Inhibiting the Pathogenic 5-HT2AR-PI3K/AKT/mTOR Signaling Axis.

Molecular neurobiology·2026
Same author

Integrated transcriptomic and proteomic analysis reveals the regulatory role of exogenous gibberellin in sugarcane internode maturation.

Frontiers in plant science·2026
Same author

Discovery of selective CDK9 degraders with anti-AML activity using PROTAC technology.

Bioorganic chemistry·2026
Same author

CGRP-mediated neuro-vascular-pulp cell crosstalk is essential for dental pulp repair.

Frontiers in cell and developmental biology·2026

Related Experiment Video

Updated: Aug 15, 2025

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

2.9K

Painless and accurate medical image analysis using deep reinforcement learning with task-oriented homogenized

Di Yuan1, Yunxin Liu1, Zhenghua Xu1

  • 1State Key Laboratory of Reliability and Intelligence of Electrical Equipment, School of Health Sciences and Biomedical Engineering, Hebei University of Technology, Tianjin, China.

Computers in Biology and Medicine
|January 5, 2023
PubMed
Summary
This summary is machine-generated.

This study introduces a deep reinforcement learning framework for automated medical image pre-processing, overcoming expert reliance and inter-institutional variability. The proposed method ensures consistent performance across diverse datasets, enhancing diagnostic accuracy.

Keywords:
Deep reinforcement learningHomogenized automatic pre-processingMedical image analysisPediatric pneumonia classificationPolicy gradient

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

478
Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

40.3K

Related Experiment Videos

Last Updated: Aug 15, 2025

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

2.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

478
Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
12:50

Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly

Published on: April 14, 2014

40.3K

Area of Science:

  • Medical Image Analysis
  • Artificial Intelligence in Healthcare
  • Deep Reinforcement Learning

Background:

  • Medical image pre-processing is crucial but often relies on expert input, hindering rapid deployment of CAD systems.
  • Inter-institutional variations in image quality and pre-processing methods lead to performance degradation and reproducibility issues.
  • Existing solutions struggle with quick deployment and consistent performance across different medical institutions.

Purpose of the Study:

  • To propose a deep-reinforcement-learning-based task-oriented homogenized automatic pre-processing (DRL-HAPre) framework.
  • To enable intelligent CAD systems to rapidly deploy and maintain satisfactory performance across diverse medical institutions.
  • To address the challenges of expert dependency and performance inconsistency in medical image pre-processing.

Main Methods:

  • Developed a DRL-HAPre framework utilizing deep reinforcement learning to automatically select optimal pre-processing operations.
  • Created a specific model, HAPre-KRS, for key region selection in pneumonia image classification.
  • Conducted extensive experiments on three pediatric pneumonia datasets with varying image qualities.

Main Results:

  • Confirmed the 'hard-to-reproduce' problem in clinical practice, linked to varying medical image qualities.
  • The HAPre-KRS model and DRL-HAPre framework significantly outperformed state-of-the-art baselines, especially on lower-quality images.
  • Homogenized pre-processing via HAPre-KRS mitigated performance degradation in cross-source applications, overcoming reproducibility issues.

Conclusions:

  • The DRL-HAPre framework effectively automates medical image pre-processing, enhancing CAD system deployment and accuracy.
  • Homogenized automatic pre-processing is essential for overcoming performance inconsistencies and reproducibility challenges in multi-institutional settings.
  • The proposed framework demonstrates significant improvements, particularly in scenarios with lower medical image quality and cross-source applications.