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

Analyzing visual-search observers using eye-tracking data for digital breast tomosynthesis images.

Zhengqiang Jiang, Mini Das, Howard C Gifford

    Journal of the Optical Society of America. A, Optics, Image Science, and Vision
    |October 17, 2017
    PubMed
    Summary

    Human gaze patterns on breast images were analyzed to improve visual-search (VS) models. The adaptive matched filter best correlated with human fixation times, enhancing VS observer design for mass detection.

    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

    Gestational lipid profile partially mediates adverse obstetric outcomes associated with polycystic ovary syndrome: a multicentre cohort study in China.

    Lipids in health and disease·2026
    Same author

    High-level expression of a newly characterized collagenase in Bacillus subtilis for efficient production of collagen peptides from fish scales.

    Food chemistry·2026
    Same author

    Biosensor-driven adaptive laboratory evolution and modular pathway engineering for high-level production of nicotinamide mononucleotide in yeast.

    Bioresource technology·2026
    Same author

    Efficient expression of γ-DL-glutamyl hydrolase in Komagataella phaffii and its application in modification of poly-γ-glutamic acid from Bacillus velezensis.

    Journal of biotechnology·2026
    Same author

    2'-Fucosyllactose Alleviates Metabolic Hypertension in Mice via Gut Microbiota Modulation and Involvement of the LPS/TLR4 Signaling.

    Journal of agricultural and food chemistry·2026
    Same author

    Computed Quantitative Planar Imaging for Targeted Alpha Therapy: Model-Based Sparse Reconstruction Validated With a Novel <sup>225</sup>Ac Epoxy Phantom.

    IEEE transactions on medical imaging·2026

    Area of Science:

    • Medical Imaging
    • Computer Vision
    • Human Factors

    Background:

    • Visual-search (VS) model observers aid in predicting human performance for detection and localization tasks.
    • Understanding human gaze characteristics is crucial for designing effective VS observers.

    Purpose of the Study:

    • To examine human gaze characteristics on breast images.
    • To inform the design of VS observers for mass detection in digital breast tomosynthesis (DBT).

    Main Methods:

    • A helmet-mounted eye-tracking system recorded human observers' gaze during mass search in 2D DBT images.
    • K-mean clustering analyzed gaze dwell times, which were compared with image morphological features.
    • Features included matched filter (MF), gradient MF, Laplacian MF, and an adaptive MF derived from a hotelling discriminant.

    Related Experiment Videos

    Main Results:

    • Correlation coefficients were computed between fixation times and extracted image features.
    • The adaptive MF demonstrated the highest correlation coefficients across DBT images of varying densities.
    • Significance tests (p-values) confirmed the correlations for five features.

    Conclusions:

    • The adaptive MF is a promising feature for improving VS observer models in breast cancer screening.
    • Gaze analysis provides valuable insights for developing more accurate and human-like VS observers.
    • This research contributes to the advancement of AI-assisted diagnostic tools in radiology.