Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025International Journal of 3D Printing Technologies and Digital IndustryOpen Access

Comparison of Machine Learning Models in Heart Failure Prediction and Their Integration Into Clinical Decision Support Systems

View Full Paper
Ask AI
Bookmark
Share

Authors

MÇMustafa Çakır

Discussion

Loading...

Member takes

Overview

Interactive machine learning application improves heart failure outcome predictions, enhancing clinical decisions.

Key Points

  • The application achieves impressive performance, with models showing test set Area Under Curve values exceeding 0.85.
  • Key clinical variables used for predictions include ejection fraction, serum creatinine, and follow-up time, demonstrating the model's practical utility.
  • Stratified cross-validation (10-fold) ensures robust evaluation of predictive accuracy, sensitivity, and specificity, confirming model effectiveness.
  • The open-source R-Shiny framework allows for real-time risk predictions and enhances model interpretability and educational resources.

Cite This Study

Mustafa Çakır (2025) studied this question.

synapsesocial.com/papers/68c182609b7b07f3a060f680https://doi.org/10.46519/ij3dptdi.1724620
View Full Paper
Ask AI
Bookmark
Share