Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025

Research on left atrial appendage thrombogenic milieu prediction model in patients with nonvalvular atrial fibrillation based on machine learning algorithm

View Full Paper
Ask AI
Bookmark
Share

Authors

LSLing SongNXNiu Xiao-qiBWBinbin Wang

Discussion

Loading...

Member takes

Overview

Machine learning effectively predicts left atrial appendage thrombogenic milieu in nonvalvular atrial fibrillation, highlighting key risk factors.

Key Points

  • The random forest model achieved the highest predictive performance for left atrial appendage thrombogenic risk.
  • Top risk factors identified include homocysteine, NT-proBNP, C-reactive protein, glycosylated hemoglobin, and ABC stroke score.
  • A total of 1217 patients were analyzed, with clear differentiation between LAATM and non-LAATM groups based on thrombus formation.
  • The machine learning techniques utilized included random forest, support vector machine, and extreme gradient boosting for model establishment.

Cite This Study

Song et al. (2025) studied this question.

synapsesocial.com/papers/68c198be9b7b07f3a061a416https://doi.org/10.21203/rs.3.rs-7301811/v1
View Full Paper
Ask AI
Bookmark
Share