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

Partitioning reference values for several subpopulations using cluster analysis.

Martin Gellerstedt1, Per Hyltoft Petersen

  • 1Department of Informatics, University of West Sweden, Trollhättan, Sweden. martin.gellerstedt@hv.se

Clinical Chemistry and Laboratory Medicine
|September 29, 2007
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

Striving to make the right decision when seeking emergency care: patients' stories of navigating support alternatives in a digital society, a qualitative study.

BMC emergency medicine·2026
Same author

Higher digital embracement is associated with lower levels of loneliness among late middle-aged and older adults.

Frontiers in public health·2026
Same author

Developing a multivariable prediction model of global health-related quality of life in patients treated for rectal cancer: a prospective study in five countries.

International journal of colorectal disease·2024
Same author

Are patients willing to take a more active role? Questionnaires to measure patients' willingness to be empowered.

Patient education and counseling·2021
Same author

Online Communities as a Driver for Patient Empowerment: Systematic Review.

Journal of medical Internet research·2021
Same author

Digitalization and Physician Learning: Individual Practice, Organizational Context, and Social Norm.

The Journal of continuing education in the health professions·2020

This study introduces a systematic clustering approach for partitioning multiple subpopulations into groups, ensuring consistent reference interval sensitivity across diverse patient characteristics. The method facilitates the creation of appropriate reference intervals for various patient groups.

Area of Science:

  • Clinical Chemistry
  • Biostatistics
  • Reference Interval Determination

Background:

  • Determining appropriate reference intervals requires assessing if distinct subpopulations necessitate separate intervals or a shared one.
  • Partitioned reference intervals enhance sensitivity across subpopulations with significant variations.
  • Existing statistical criteria primarily address two subpopulations, with recent advancements for multiple groups.

Purpose of the Study:

  • To develop a systematic approach for clustering multiple subpopulations for reference interval determination.
  • To address the lack of formal methods for grouping subpopulations that may share common reference intervals.
  • To harmonize reference interval sensitivity across diverse patient characteristics.

Main Methods:

Related Experiment Videos

  • Applied a clustering technique to data comprising multiple subpopulations.
  • Measured distances between reference limits to successively pool subpopulations with short distances.
  • Defined clusters as groups of close subpopulations, pooling new members if partitioning criteria are met.

    • Main Results:

    • A formalized procedure for partitioning Gaussian (or Gaussian-transformable) subpopulations into clusters was developed.
    • The approach enables the identification of an optimal number of groups and corresponding reference intervals.
    • The method was easily implemented using a dedicated computer program.

    Conclusions:

    • The proposed clustering procedure offers a systematic method for analyzing multiple subpopulations.
    • This is the first formalized approach to guide the grouping of subpopulations for reference interval analysis.
    • The methodology simplifies the process of establishing appropriate reference intervals for diverse patient cohorts.