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: Oct 4, 2025

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing
06:25

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Published on: February 23, 2024

750

Deep learning versus the human visual system for detecting motion blur in radiography.

Rie Tanaka1, Shiho Nozaki2, Futa Goshima2

  • 1Kanazawa University, College of Medical, Pharmaceutical and Health Sciences, Kanazawa, Japan.

Journal of Medical Imaging (Bellingham, Wash.)
|February 2, 2022
PubMed
Summary

Related Concept Videos

Visual System01:26

Visual System

781
Light enters the eye through the cornea, a transparent, dome-shaped surface covering the surface of the eyeball that helps to direct and focus incoming light. This light is then channeled toward the pupil, an adjustable opening whose size is controlled by the iris. The iris, a pigmented muscle, regulates the amount of light entering the eye by contracting or dilating the pupil, thereby ensuring optimal light levels for clear vision.
Once through the pupil, the light passes through the lens, a...
781

You might also read

Related Articles

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

Sort by
Same author

A reproducible framework for constructing a longitudinal low-dose CT screening image database: implementation using 11 years of real-world data.

Radiological physics and technology·2026
Same author

Optimization of pulmonary blood flow analysis method using dynamic chest radiography.

Journal of applied clinical medical physics·2026
Same author

Historical and clinical contributions of the JSRT chest radiograph database to medical imaging research.

Radiological physics and technology·2026
Same author

Deep learning-based lung volume estimation with dynamic chest radiography.

Journal of applied clinical medical physics·2026
Same author

Furanyl bis(indolyl)methane as a palladium ion-selective chromogenic agent.

Organic & biomolecular chemistry·2024
Same author

Impact of multiple H/D replacements on the physicochemical properties of flurbiprofen.

RSC medicinal chemistry·2023

Deep learning models accurately detect subtle motion blur in digital radiographs, outperforming human observers on preview monitors. This AI approach enhances diagnostic accuracy by identifying issues missed during initial reviews.

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Image quality is crucial for accurate radiographic diagnosis.
  • Blurring in radiographs can be subtle and often missed on initial review monitors.
  • This can lead to the necessity of image retakes, increasing costs and delaying patient care.

Purpose of the Study:

  • To compare the performance of a deep learning approach with human observers in detecting motion blur in digital radiographs.
  • To evaluate the effectiveness of deep convolution neural networks (DCNNs) in identifying subtle image blurring.
  • To assess the potential of AI in improving the initial review of radiographic images.

Main Methods:

  • A dataset of 99 radiographs (57 blurry, 42 non-blurry) was used.
Keywords:
LCD monitorROC studydeep learningmotion blurradiography

Related Experiment Videos

Last Updated: Oct 4, 2025

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing
06:25

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Published on: February 23, 2024

750
  • Six human observers rated images on preview and diagnostic liquid crystal displays (LCDs).
  • A deep convolution neural network (DCNN) was trained and tested using ninefold cross-validation.

    • Main Results:

    • The DCNN achieved an average area under the ROC curve (AUC) of 0.955, surpassing human observers (0.827 on preview LCDs, 0.947 on diagnostic LCDs).
    • The DCNN demonstrated high performance (sensitivity 94.8%, specificity 96.8%, accuracy 95.6%) in detecting motion blur.
    • The DCNN identified slight motion blur missed by human observers using preview LCDs.

    Conclusions:

    • Deep learning models effectively detect subtle motion blur in digital radiographs, even when unnoticeable on preview monitors.
    • AI-powered blur detection can assist the human visual system in the initial review of radiographs.
    • This technology has the potential to reduce unnecessary image retakes and improve diagnostic efficiency.