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

Impact of data encoding and thyroids dysfunctions.

Karina Gibert1, Zdenko Sonicki, Juan Carlos Martín

  • 1Dept. of Statistics and Operation Research, UPC, C. Pau Gargallo, 5, Barcelona. karina@eio.upc.es

Studies in Health Technology and Informatics
|October 6, 2004
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

Regional disparities in subjective wellbeing across Europe: A fuzzy hybrid TOPSIS approach.

PloS one·2026
Same author

Editorial: Sport practice and physical activity-the social function of sport in contemporary societies.

Frontiers in sports and active living·2025
Same author

Measuring the physical activity in the EU using a fuzzy hybrid synthetic index and an ordered probit model.

Frontiers in sports and active living·2025
Same author

Optimizing Engagement: Factors Influencing Family Participation in a Positive Parenting Program among Vulnerable Households with Young Children.

Psychosocial intervention·2025
Same author

Expression of Wnt signaling proteins LEF1, β-catenin, GSK3β, DVL1, and N-myc varies across retinoblastoma subtypes and pRb phosphorylation status.

Scientific reports·2024
Same author

A Novel Time-Aware Deep Learning Model Predicting Myopia in Children and Adolescents.

Ophthalmology science·2024
Same journal

A GenAI Pipeline for Violinist Kinematic Data Management.

Studies in health technology and informatics·2026
Same journal

AMAL-For-Qatar: A Comprehensive AI Ecosystem for Fetal Ultrasound Analysis - Project Overview and Achievements.

Studies in health technology and informatics·2026
Same journal

Longitudinal Treatment-Aware Multimodal AI for Dermatology: A Scoping Review.

Studies in health technology and informatics·2026
Same journal

Predicting Postpartum Depression Using Imbalance-Aware Machine Learning.

Studies in health technology and informatics·2026
Same journal

Validation of Deep-Learning Models for Autosegmentation of Brain Metastases.

Studies in health technology and informatics·2026
Same journal

Delay-Dependent Gating in Modular RNNs.

Studies in health technology and informatics·2026
See all related articles

Preprocessing numerical hormone levels into categories impacts clustering results for thyroid dysfunction analysis. This study examines how this common medical practice affects data grouping in patient datasets.

Area of Science:

  • Medical Informatics
  • Data Science
  • Endocrinology

Background:

  • Medical data often involves encoding numerical variables (e.g., hormone levels) into categories (low, normal, high) based on clinical criteria.
  • Clustering algorithms rely on distance metrics, and the nature of input data significantly influences grouping outcomes.
  • Thyroid dysfunction analysis frequently employs categorical representations of hormone levels like triiodothyronine (T3), thyroxine (T4), and thyroid-stimulating hormone (TSH).

Purpose of the Study:

  • To investigate the impact of transforming numerical hormone levels into categorical labels on clustering results.
  • To evaluate how this common medical data preprocessing technique affects patient data analysis in the context of thyroid disorders.

Main Methods:

  • Utilized a real-world dataset of patients from a hospital in Zagreb, Croatia.

Related Experiment Videos

  • Applied clustering algorithms to both raw numerical hormone data and preprocessed categorical hormone data.
  • Compared the clustering outcomes derived from the two data representations.

    • Main Results:

    • The conversion of numerical hormone levels (T3, T4, TSH) to categorical labels alters the results of clustering analyses.
    • Different distance/similarity coefficients used in clustering yield varying outcomes depending on the data's nature (numerical vs. categorical).

    Conclusions:

    • The preprocessing of numerical hormone data into categorical labels is a critical step that can significantly influence the interpretation of clustering results in medical research.
    • Clinicians and researchers should be aware of how data transformation affects clustering outcomes, especially in the study of endocrine disorders like thyroid dysfunction.