Automated triage in pandemics: Support vector machine for efficient patient classification

doi:10.1016/j.idnow.2026.105267

Automated triage in pandemics: Support vector machine for efficient patient classification

2026-04-03

0

PRE

AI

A

Ana Gabriela Gallardo-Hernández

T

Tania Colín-Martínez

M

Marcos A. González-Olvera

Eli Abraham Vázquez-Orduña

DOI：10.1016/j.idnow.2026.105267

Original

Original request for help

Save

Abstract

En 中文

• Before the COVID-19 pandemic emergency departments around the world were already overcrowded. • Even when conventional triage protocols are well established, there is room for improvement, and it is important to recognize that physicians can always override the triage algorithm based on their experience to prevent under-triage. • Based on support vector machines (SVM), the system designed in this study can not only replicate established triage decisions but also the physician-driven criteria used to prevent under-triage. • SVM system can identify with 94% accuracy green and red patients using only 96 patients for training. This system can be used to replicate a triage system for any new threat-causing pathogen. • In regular operation of triage systems, SVM systems can be designed to consider not only the respiratory systems specific to COVID-19, but all types of emergencies covered by conventional triage algorithms.

Keywords:

COVID-19

Pandemics

Resource allocation

InfectiousDiseases

SupportVector Machines

Automated triage in pandemics: Support vector machine for efficient patient classification

Abstract

Journal

Organization