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 Videos

Recurrence quantification analysis of radiologists' scanpaths when interpreting mammograms.

Ziba Gandomkar1, Kevin Tay2, Patrick C Brennan1

  • 1Image Optimisation and Perception Group (MIOPeG), Discipline of Medical Imaging and Radiation Sciences, The University of Sydney, Sydney, NSW, Australia.

Medical Physics
|April 26, 2018
PubMed
Summary

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

A novel deep learning-based grading system for assessing breast arterial calcification on mammograms, as an independent risk factor for predicting adverse cardiovascular events.

La Radiologia medica·2026
Same author

Initiators of Semaglutide in General Practice in New South Wales, 2020-2023: A Retrospective Cohort Study.

Heart, lung & circulation·2026
Same author

Amplifying image quality gain in x-ray phase contrast imaging of mastectomy samples with deep learning denoising.

Physics in medicine and biology·2026
Same author

Semi-Supervised Deep Learning-Based Model for Segmentation of Breast Arterial Calcification on Screening Mammograms.

Canadian Association of Radiologists journal = Journal l'Association canadienne des radiologistes·2025
Same author

Factors associated with junior doctor plain trauma X-ray interpretation accuracy and strategies for improvement: a scoping review.

BMC medical imaging·2025
Same author

Accuracy of junior doctor plain trauma X-ray interpretation: a systematic review and meta-analysis.

BMC medical imaging·2025

Recurrence quantification analysis (RQA) metrics effectively distinguish expert radiologists' mammogram reading patterns from less-experienced readers. These metrics also reveal scanpath dynamics related to case complexity and radiologist decisions.

Area of Science:

  • Medical Imaging
  • Radiology
  • Cognitive Science

Background:

  • Expertise in mammography interpretation is crucial for accurate cancer detection.
  • Understanding the visual search strategies of radiologists can improve training and diagnostic performance.
  • Recurrence quantification analysis (RQA) offers a novel approach to quantify complex temporal patterns in eye-tracking data.

Purpose of the Study:

  • To develop a classifier using RQA metrics to differentiate expert radiologists' scanpaths from those of less-experienced readers.
  • To investigate how spatiotemporal dynamics in mammographic scanpaths relate to case characteristics and radiologist expertise using RQA.

Main Methods:

  • Eye movements of 8 radiologists (4 experienced, 4 less-experienced) were recorded while viewing 120 mammograms.
Keywords:
breast cancereye movementsmammogramrecurrence quantification analysisscanpaths

Related Experiment Videos

  • Ten RQA measures were extracted to analyze fixation patterns, including recurrent, laminar, and deterministic eye movements.
  • A classifier was built using RQA and conventional eye-tracking parameters, validated with leave-one-out cross-validation.

    • Main Results:

    • All RQA measures significantly differed between experienced and less-experienced readers.
    • The RQA-based classifier achieved an area under the ROC curve of 0.89 for distinguishing expert scanpaths.
    • Experienced radiologists showed more refixations and laminar/deterministic sequences in lesion areas.
    • RQA measures correlated with case pathology and radiologist decisions, particularly for experienced readers.

    Conclusions:

    • RQA metrics provide a quantitative method to differentiate between experienced and less-experienced radiologists' visual search patterns.
    • RQA analysis reveals that scanpath dynamics are influenced by case complexity and individual radiologist decision-making.