Machine learning models for predicting in-hospital mortality in patient with sepsis: Analysis of vital sign dynamics

PurposeTo build machine learning models for predicting the risk of in-hospital death in patients with sepsis within 48 h, using only dynamic changes in the patient's vital signs.MethodsThis retrospective observational cohort study enrolled septic patients from five emergency departments (ED) in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Frontiers in medicine Jg. 9; S. 964667
Hauptverfasser: Cheng, Chi-Yung, Kung, Chia-Te, Chen, Fu-Cheng, Chiu, I-Min, Lin, Chun-Hung Richard, Chu, Chun-Chieh, Kung, Chien Feng, Su, Chih-Min
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Lausanne Frontiers Media SA 20.10.2022
Frontiers Media S.A
Schlagworte:
Algorithms
Antibiotics
Artificial intelligence
Blood pressure
Cardiac arrest
dynamic vital sign
Electronic health records
Hospitals
Machine learning
machine learning model
Medicine
Mortality
Neural networks
Patients
prediction model
Sepsis
Variables
Vital signs
ISSN:2296-858X, 2296-858X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PurposeTo build machine learning models for predicting the risk of in-hospital death in patients with sepsis within 48 h, using only dynamic changes in the patient's vital signs.MethodsThis retrospective observational cohort study enrolled septic patients from five emergency departments (ED) in Taiwan. We adopted seven variables, i.e., age, sex, systolic blood pressure, diastolic blood pressure, heart rate, respiratory rate, and body temperature.ResultsAmong all 353,253 visits, after excluding 159,607 visits (45%), the study group consisted of 193,646 ED visits. With a leading time of 6 h, the convolutional neural networks (CNNs), long short-term memory (LSTM), and random forest (RF) had accuracy rates of 0.905, 0.817, and 0.835, respectively, and the area under the receiver operating characteristic curve (AUC) was 0.840, 0.761, and 0.770, respectively. With a leading time of 48 h, the CNN, LSTM, and RF achieved accuracy rates of 0.828, 0759, and 0.805, respectively, and an AUC of 0.811, 0.734, and 0.776, respectively.ConclusionBy analyzing dynamic vital sign data, machine learning models can predict mortality in septic patients within 6 to 48 h of admission. The performance of the testing models is more accurate if the lead time is closer to the event.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Edited by: Silvia Spoto, Policlinico Universitario Campus Bio-Medico, Italy
Reviewed by: Domenica Marika Lupoi, Policlinico Universitario Campus Bio-Medico, Italy; Rishikesan Kamaleswaran, Emory University, United States
This article was submitted to Infectious Diseases – Surveillance, Prevention and Treatment, a section of the journal Frontiers in Medicine
ISSN:2296-858X
2296-858X
DOI:10.3389/fmed.2022.964667