Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025Scientific ReportsOpen Access

Machine learning to predict bacteriuria in the emergency department

View Full Paper
Ask AI
Bookmark
Share

Authors

JSJohnathan M. SheeleRCRonna L. CampbellDJDerick D. Jones

Discussion

Loading...

Member takes

Overview

Retrospective analysis demonstrates machine learning predicts urine culture outcomes in patients, suggesting enhanced treatment decisions.

Key Points

  • XGBoost accurately predicted bacteriuria, showing an AUROC of 93.1% for ≥100,000 CFU/mL outcomes.
  • Using data from over 62,000 emergency department encounters, machine learning models were assessed against traditional diagnoses.
  • Logistic regression, k-nearest neighbors, random forest, and deep neural networks were used to evaluate predictive performance.
  • Findings indicate that machine learning could improve clinical decision-making regarding UTI treatment in emergency settings.

Cite This Study

Sheele et al. (2025) studied this question.

synapsesocial.com/papers/68c1d03e54b1d3bfb60f6f8fhttps://doi.org/10.1038/s41598-025-16677-z
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. 1Interpretable Machine Learning Models for Predicting Critical Outcomes in Patients with Suspected Urinary Tract Infection with Positive Urine Culture2024 · 16 citations
  2. 2Machine Learning Algorithms for Predicting Urinary Tract Infections: Integration of Demographic Data and Dipstick Reflectance Results2025 · 6 citations
  3. 3Survey on Machine Learning Models to Analyze Urinary Tract Infection Data2024 · 1 citations
  4. 4P-780. Machine Learning Models for Early Urinary Tract Infection (UTI) Prediction from Electronic Health Records2026 · 1 citations
  5. 5Machine learning for the prediction of urosepsis using electronic health record data2024