Machine learning-based model to classify Emergency Severity Index (ESI) levels 1-3 in febrile patients with tachycardia: Thailand Triage Prediction System (TTPS) Triage
Conditions
Interventions
Patients classified as Emergency Severity Index (ESI) level 1 by expert consensus at triage.,Patients classified as Emergency Severity Index (ESI) level 2 by expert consensus at triage.,Patients class
Diagnostic,Diagnostic,Diagnostic
ESI Level 1,ESI Level 2,ESI Level 3
Sponsors
Lampang Hospital, Thailand
Eligibility
Sex/Gender
All
Age
18 Years to No maximum
Inclusion criteria
Inclusion criteria: 1. Adults aged 18 years or older 2. Fever defined as body temperature 37.6 degrees Celsius or higher measured using a non contact infrared thermometer 3. Pulse rate greater than 100 beats per minute
Exclusion criteria
Exclusion criteria: 1. Patients referred from other hospitals 2. Patients with incomplete data 3. Patients with an inconclusive outcome in the expert evaluation of Emergency Severity Index (ESI) level
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Predicted Emergency Severity Index (ESI) levels 1-3 At the triage area in the emergency department Triage classification determined by machine learning-based models (Random Forest, XGBoost, Gradient Boosting), compared against expert consensus using ESI version 5. | — |
Secondary
| Measure | Time frame |
|---|---|
| Model performance (AUROC, accuracy, recall, F1-score, precision) After model evaluation in the internal validation phase Computed from predictions on the held-out test dataset of 100 patients | — |
Countries
Thailand
Contacts
Public ContactChanitda Wicha
Department of Emergency Medicine, Lampang Hospital
Outcome results
None listed