Screening of Pneumonia and Urinary Tract Infection at Triage using TriNet

TL;DR

TriNet is a machine learning model designed to automate triage screening for pneumonia and urinary tract infections, improving accuracy and efficiency in emergency departments by reducing over-testing and wait times.

Contribution

This paper introduces TriNet, a novel machine learning approach that outperforms existing benchmarks in triage screening for common infections, streamlining emergency department workflows.

Findings

Achieved high positive predictive values: 0.86 for pneumonia and 0.93 for urinary tract infection.

Outperforms current clinical benchmarks in screening accuracy.

Offers a cost-free, non-invasive method to improve emergency department efficiency.

Abstract

Due to the steady rise in population demographics and longevity, emergency department visits are increasing across North America. As more patients visit the emergency department, traditional clinical workflows become overloaded and inefficient, leading to prolonged wait-times and reduced healthcare quality. One of such workflows is the triage medical directive, impeded by limited human workload, inaccurate diagnoses and invasive over-testing. To address this issue, we propose TriNet: a machine learning model for medical directives that automates first-line screening at triage for conditions requiring downstream testing for diagnosis confirmation. To verify screening potential, TriNet was trained on hospital triage data and achieved high positive predictive values in detecting pneumonia (0.86) and urinary tract infection (0.93). These models outperform current clinical benchmarks, indicating that machine-learning medical directives can offer cost-free, non-invasive screening with high specificity for common conditions, reducing the risk of over-testing while increasing emergency department efficiency.