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 Experiment Video

Updated: Nov 25, 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

3.2K

A Hybrid Deep Learning Approach for Spatial Trigger Extraction from Radiology Reports.

Surabhi Datta1, Kirk Roberts1

  • 1School of Biomedical Informatics, University of Texas Health Science Center at Houston Houston TX, USA.

Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing
|December 18, 2020
PubMed
Summary

This study introduces a hybrid deep learning method to automatically identify spatial expressions in radiology reports. The approach significantly improves the accuracy of pinpointing anatomical relationships and findings.

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

Synthesized annotation guidelines are knowledge-lite boosters for clinical information extraction.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Education Research: Integration of Trainee and Faculty Clinics at an Academic Medical Center: Improving Quality of Care and Education.

Neurology. Education·2026
Same author

AI-Assisted Systematic Literature Review of the Economic Burden of Pneumococcal Disease: Development and Validation Study.

JMIR AI·2026
Same author

Hootation: A GUI and API library for ontology validation and verbalization.

Proceedings. IEEE International Conference on Semantic Computing·2026
Same author

Clinical document metadata extraction: A scoping review.

Journal of biomedical informatics·2026
Same author

Facilitating Clinical Information Extraction with Synthetic Data and Ontology using Large Language Models.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2026

Area of Science:

  • Natural Language Processing (NLP)
  • Medical Imaging Informatics
  • Artificial Intelligence in Medicine

Background:

  • Radiology reports contain crucial clinical data, often described using spatial expressions.
  • Spatial expressions define the relationships between radiographic findings, medical devices, and anatomical structures.
  • The complexity and variability of these expressions pose challenges for automated analysis.

Purpose of the Study:

  • To develop and evaluate a hybrid deep learning NLP method for automatically identifying spatial expression terms.
  • To enhance the extraction of clinically relevant spatial information from diverse radiology sub-domains.
  • To improve upon existing methods for spatial trigger detection in medical text.

Main Methods:

  • A hybrid approach combining exact matching, domain-specific constraints, and a BERT-based classifier.

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

817

Related Experiment Videos

Last Updated: Nov 25, 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

3.2K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

817
  • Candidate spatial triggers were generated via exact match with known terms.
  • Domain-specific constraints were applied for filtering, followed by BERT classification for prediction.

    • Main Results:

    • The proposed hybrid method demonstrated promising results in identifying spatial expression terms.
    • Achieved a significant improvement of 24 points in the average F1 measure compared to a standard BERT sequence labeler.
    • Successfully applied to three different radiology sub-domains.

    Conclusions:

    • The hybrid deep learning NLP method is effective for automated spatial expression identification in radiology.
    • This approach offers a substantial advancement over standard sequence labeling techniques for this task.
    • The findings suggest improved clinical information extraction from radiology reports through enhanced NLP capabilities.