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

Estimation and prediction for cancer screening models using deconvolution and smoothing.

P F Pinsky1

  • 1Division of Cancer Prevention, National Cancer Institute, Bethesda, Maryland 20892, USA. pp4f@nih.gov

Biometrics
|June 21, 2001
PubMed
Summary
This summary is machine-generated.

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

Active follow-up versus passive linkage with cancer registries for case ascertainment in a cohort.

Cancer epidemiology·2016
Same author

Evidence of a healthy volunteer effect in the prostate, lung, colorectal, and ovarian cancer screening trial.

American journal of epidemiology·2007
Same author

Relationship of demographic and clinical factors to free and total prostate-specific antigen.

Urology·2001
Same author

A multi-stage model of adenoma development.

Journal of theoretical biology·2000
Same author

Prognostic factors for hematologic cancers.

Hematology/oncology clinics of North America·2000
Same author

Lung cancer among workers in chromium chemical production.

American journal of industrial medicine·2000

This study introduces a new method for cancer screening analysis by assuming known underlying cancer incidence. This approach refines preclinical incidence estimation, improving accuracy for sojourn time and lead time calculations.

Area of Science:

  • Epidemiology
  • Biostatistics
  • Mathematical Modeling

Background:

  • Cancer screening trials commonly use a convolution model relating incidence to preclinical phase duration.
  • Estimating preclinical incidence (g) and screening sensitivity requires fitting parameters, often with prior assumptions about g.

Purpose of the Study:

  • To develop a more accurate method for estimating cancer screening trial parameters by assuming known underlying incidence (I).
  • To improve estimates of sojourn time, lead time, and overdiagnosis by refining the preclinical incidence function (g).

Main Methods:

  • Utilized a convolution model where cancer incidence (I) is the convolution of preclinical incidence (g) and preclinical phase time density (f).
  • Assumed underlying incidence (I) is known, allowing preclinical incidence (g) to be solved via numerical deconvolution of f.

Related Experiment Videos

  • Incorporated a smoothing procedure to stabilize deconvolution and ensure a realistic g function.
  • Integrated competing mortality into the convolution model.

    • Main Results:

    • Eliminated the need to estimate or assume preclinical incidence (g), simplifying model fitting.
    • Achieved a realistic preclinical incidence function (g) that accurately reproduces the original incidence (I) upon convolution.
    • Demonstrated more accurate estimation of sojourn time and lead time.
    • Enabled the estimation of quantities related to overdiagnosis.

    Conclusions:

    • The new method provides more accurate parameter estimations in cancer screening trials.
    • Incorporating known underlying incidence and competing mortality enhances the reliability of sojourn time and lead time estimates.
    • The refined model facilitates a better understanding of overdiagnosis in cancer screening.