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

Interval-coded texture features for artifact rejection in automated cervical cytology.

J H Tucker1, K Rodenacker, U Juetting

  • 1Clinical and Population Cytogenetics Unit, Medical Research Council, Edinburgh, Scotland.

Cytometry
|September 1, 1988
PubMed
Summary

This study improved cervical cancer screening by developing new texture features to better distinguish abnormal cells from artifacts in automated prescreening systems.

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 eureka moment - using coalface data for discovery on TB disease and financial relief.

IJTLD open·2026
Same author

TB in pregnancy: a 5-year review and future recommendations in Victoria, Australia.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2026
Same author

Epidemiological Society: On the Use of Vegetable and Mineral Acids in the Treatment, Prophylactic and Remedial, of Epidemic Disorders of the Bowels.

The North-Western medical and surgical journal·2023
Same author

Risk factors for TB in Australia and their association with delayed treatment completion.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2022
Same author

Multigene phylogenetics of euglenids based on single-cell transcriptomics of diverse phagotrophs.

Molecular phylogenetics and evolution·2021
Same author

Use of Artificial Intelligence to understand adults' thoughts and behaviours relating to COVID-19.

Perspectives in public health·2021

Area of Science:

  • Digital pathology
  • Medical image analysis
  • Cytopathology

Background:

  • Automated cervical cytology prescreening systems like CERVIFIP face challenges in differentiating abnormal cells from noncellular artifacts.
  • Accurate cell classification is crucial for reliable cervical cancer diagnosis and patient management.

Purpose of the Study:

  • To enhance the separation of abnormal cells from artifacts in the CERVIFIP system by investigating novel object texture features.
  • To improve the accuracy of automated cervical cytology prescreening.

Main Methods:

  • Investigated 22 statistical object texture features derived from pixel density histograms and filtered image regions.
  • Calculated features for 231 images (100 cells, 131 artifacts) identified as 'Suspect Cells' by CERVIFIP.

Related Experiment Videos

  • Tested features using hierarchical and linear discriminant classifiers, including maximum likelihood discrimination.

    • Main Results:

    • A hierarchical classifier using the two best features achieved 83% correct classification.
    • One selected feature effectively removed poorly focused objects.
    • Maximum likelihood discrimination with all 22 features yielded a 90% overall correct classification rate.

    Conclusions:

    • Object texture features significantly improve the discrimination between abnormal cells and artifacts in automated cervical cytology.
    • The developed features enhance the performance of prescreening systems, potentially leading to more accurate diagnoses.
    • Further integration of these features can optimize automated cervical cancer detection systems.