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

Four new spider species of the <i>Pholcus phungiformes</i> species group (Araneae, Pholcidae) from South Korea.

ZooKeys·2026
Same author

Biomarkers.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Basic Science and Pathogenesis.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Rapid antidepressant effect of single-bout exercise is mediated by adiponectin-induced APPL1 nucleus translocation in anterior cingulate cortex.

Molecular psychiatry·2025
Same author

Revision of the velvet spiders (Araneae, Eresidae) with a new record from South Korea.

Biodiversity data journal·2025
Same author

PA-Win2: In Silico-Based Discovery of a Novel Peptide with Dual Antibacterial and Anti-Biofilm Activity.

Antibiotics (Basel, Switzerland)·2025

Related Experiment Video

Updated: Sep 16, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

692

Radiology report generation using automatic keyword adaptation, frequency-based multi-label classification and

Zebang He1, Alex Ngai Nick Wong2, Jung Sun Yoo1

  • 1Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong Special Administrative Region of China.

Computers in Biology and Medicine
|July 4, 2025
PubMed
Summary

This study introduces a novel deep learning framework for automated radiology report generation, enhancing explainability and accuracy. The method uses transparent keywords and large language models to create reliable, human-like reports, improving clinical workflows.

Keywords:
Automatic keyword adaptationFrequency-based multi-label classificationLarge language modelRadiology report generation

More Related Videos

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

267
Cloud-Based Phrase Mining and Analysis of User-Defined Phrase-Category Association in Biomedical Publications
09:20

Cloud-Based Phrase Mining and Analysis of User-Defined Phrase-Category Association in Biomedical Publications

Published on: February 23, 2019

8.9K

Related Experiment Videos

Last Updated: Sep 16, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

692
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
05:33

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

267
Cloud-Based Phrase Mining and Analysis of User-Defined Phrase-Category Association in Biomedical Publications
09:20

Cloud-Based Phrase Mining and Analysis of User-Defined Phrase-Category Association in Biomedical Publications

Published on: February 23, 2019

8.9K

Area of Science:

  • Artificial Intelligence in Medical Imaging
  • Natural Language Processing for Healthcare

Background:

  • Manual radiology report generation is time-consuming and prone to delays.
  • Existing deep learning methods lack explainability and adaptability.

Purpose of the Study:

  • To develop a novel deep learning framework for explainable and accurate radiology report generation.
  • To improve adaptability of AI models to diverse clinical settings.

Main Methods:

  • Replaced black-box features with transparent keyword lists using multi-label classification.
  • Implemented automatic keyword adaptation for dynamic configuration.
  • Utilized a frequency-based strategy for keyword imbalance.
  • Employed a pre-trained text-to-text large language model (LLM) for report synthesis.

Main Results:

  • Achieved superior performance over state-of-the-art methods on IU-XRay and MIMIC-CXR datasets.
  • Demonstrated enhanced explainability, accuracy, and reliability in report generation.
  • Ablation studies confirmed the robustness and effectiveness of individual framework components.

Conclusions:

  • Developed a novel deep learning method for high-quality, explainable radiology reports.
  • The framework addresses explainability gaps and offers a flexible automated pipeline.
  • Aimed to reduce radiologist workload and support human-AI interaction.