Machine Learning Analysis of Biomarkers and Infectious Sites in Elderly Sepsis: Distinguishing Escherichia coli from Non-Escherichia coli Infections with a Random Forest Model DOI Creative Commons

Bu-Ren Li,

Ying Zhuo,

Shi‐Yan Zhang

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: March 6, 2024

Abstract This study examines the challenge of accurately diagnosing sepsis subtypes in elderly patients, focusing on distinguishing between Escherichia coli and non-E. infections. Utilizing machine learning, we conducted a retrospective analysis 119 employing Random Forest model to evaluate clinical biomarkers infection sites. The demonstrated high diagnostic accuracy, with an overall accuracy 87.5%, impressive precision recall rates 93.3% respectively. It identified site, Platelet Distribution Width (PDW), platelet count, Procalcitonin (PCT) levels as key predictors, while logistic regression underscored significance smoking. Achieving F1 Score 90.3% ROC AUC 88.0%, our effectively differentiates subtypes. methodology offers potential for enhancing diagnosis, improving patient outcomes, contributing advancement medicine field infectious diseases.

Language: Английский

Random forest differentiation of Escherichia coli in elderly sepsis using biomarkers and infectious sites DOI Creative Commons

Bu-Ren Li,

Ying Zhuo,

Yingying Jiang

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: June 5, 2024

Abstract This study addresses the challenge of accurately diagnosing sepsis subtypes in elderly patients, particularly distinguishing between Escherichia coli (E. coli) and non- E. infections. Utilizing machine learning, we conducted a retrospective analysis 119 employing random forest model to evaluate clinical biomarkers infection sites. The demonstrated high diagnostic accuracy, with an overall accuracy 87.5%, impressive precision recall rates 93.3% respectively. It identified sites, platelet distribution width, reduced count, procalcitonin levels as key predictors. achieved F1 Score 90.3% area under receiver operating characteristic curve 88.0%, effectively differentiating subtypes. Similarly, logistic regression least absolute shrinkage selection operator underscored significance infectious methodology shows promise for enhancing diagnosis contributing advancement medicine field diseases.

Language: Английский

Citations

4
A microbiological and genomic perspective of globally collected Escherichia coli from adults hospitalized with invasive E. coli disease DOI Creative Commons
Enya Arconada Nuin, Tuba Vilken, Basil Britto Xavier

et al.

Journal of Antimicrobial Chemotherapy, Journal Year: 2024, Volume and Issue: 79(9), P. 2142 - 2151

Published: July 13, 2024

Escherichia coli can cause infections in the urinary tract and normally sterile body sites leading to invasive E. disease (IED), including bacteraemia sepsis, with older populations at increased risk. We aimed estimate theoretical coverage rate by ExPEC4V 9V vaccine candidates. In addition, we better understanding diversity of isolates, their genetic phenotypic antimicrobial resistance (AMR), sequence types (STs), O-serotypes bacterial population structure.

Language: Английский

Citations

3
Distribution of virulence genes and antimicrobial resistance of Escherichia coli isolated from hospitalized neonates: A multi-center study across China DOI Creative Commons
Yuting Guo,

Ruiqi Xiao,

Jinxing Feng

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(16), P. e35991 - e35991

Published: Aug. 1, 2024

is the most common gram-negative pathogen to cause neonatal infections. Contemporary virulence characterization and antimicrobial resistance (AMR) data of

Language: Английский

Citations

2
Machine Learning Analysis of Biomarkers and Infectious Sites in Elderly Sepsis: Distinguishing Escherichia coli from Non-Escherichia coli Infections with a Random Forest Model DOI Creative Commons

Bu-Ren Li,

Ying Zhuo,

Shi‐Yan Zhang

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: March 6, 2024

Abstract This study examines the challenge of accurately diagnosing sepsis subtypes in elderly patients, focusing on distinguishing between Escherichia coli and non-E. infections. Utilizing machine learning, we conducted a retrospective analysis 119 employing Random Forest model to evaluate clinical biomarkers infection sites. The demonstrated high diagnostic accuracy, with an overall accuracy 87.5%, impressive precision recall rates 93.3% respectively. It identified site, Platelet Distribution Width (PDW), platelet count, Procalcitonin (PCT) levels as key predictors, while logistic regression underscored significance smoking. Achieving F1 Score 90.3% ROC AUC 88.0%, our effectively differentiates subtypes. methodology offers potential for enhancing diagnosis, improving patient outcomes, contributing advancement medicine field infectious diseases.

Language: Английский

Citations

0