Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 14, 2018BioData MiningOpen Access

Evolutionary methods for variable selection in the epidemiological modeling of cardiovascular diseases

View Full Paper
Ask AI
Bookmark
Share

Population

1465 study subjects from the epidemiological follow-up study KIHD with 433 predictor variables

Comparison

Evolutionary variable selection method combined… vs Conventional stepwise selection

Design

Other

Authors

CBChristina BresterJKJussi KauhanenTTTomi‐Pekka Tuomainen

Discussion

Loading...

Member takes

Overview

Reduces feature burden for interpretable CVD prediction models; leaves open external validation before clinical use.

Structured PICO

P
Population
1465 study subjects from the epidemiological follow-up study KIHD (Kuopio Ischemic Heart Disease Risk Factor Study) with 433 predictor variables
I
Intervention
Evolutionary variable selection method combined with Generalized Linear Logistic Regression and Support Vector Machine
C
Comparator
Conventional stepwise selection
O
Outcome
Predictive ability of the models evaluated with an F-score metric for incidents of cardiovascular diseases

An evolutionary variable selection method effectively reduced the number of predictors from 433 to 38 without compromising the predictive performance for cardiovascular disease incidence.

Cite This Study

Brester et al. (2018) studied this question.

synapsesocial.com/papers/6a85661335bc0eada8d59afdhttps://doi.org/10.1186/s13040-018-0180-x
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Modeling and variable selection in epidemiologic analysis.1989 · 2,245 citations
  2. 2Evaluating variable selection methods for multivariable regression models: A simulation study protocol2024 · 44 citations
  3. 3Evaluating Feature Selection Methods and Feature Contributions for Cardiovascular Disease Risk Prediction2025
  4. 4Variable selection for clinical prediction models in low-dimensional data - a simulation study comparing traditional regression and machine learning methods2026 · 1 citations
  5. 5Comparison of parametric and machine methods for variable selection in simulated Genetic Analysis Workshop 19 data2016 · 3 citations