Machine learning early prediction of respiratory syncytial virus in pediatric hospitalized patients

Area of Science:

• Pediatric infectious diseases
• Medical informatics
• Machine learning in healthcare

Background:

• Respiratory syncytial virus (RSV) is a major cause of pediatric illness, leading to millions of infections annually in the US.
• Undetected RSV infections can result in outbreaks, posing risks to other hospitalized patients.
• Current methods for predicting RSV in hospitalized children are limited, lacking rapid and reliable tools beyond diagnostic testing.

Purpose of the Study:

• To develop and validate a machine learning model for predicting RSV positivity in pediatric inpatients.
• To provide clinicians with a rapid and cost-effective tool for identifying potential RSV cases.

Main Methods:

• A retrospective study utilizing pediatric electronic health record (EHR) data.
• Development of a machine learning model to predict RSV infection based on EHR data.
• Evaluation of model performance using area under the receiver-operating curve (AUC), sensitivity, and specificity.

Main Results:

• The machine learning model achieved excellent predictive performance.
• Demonstrated an AUC of 0.919, indicating strong discrimination.
• Achieved a sensitivity of 0.802 and a specificity of 0.876.

Conclusions:

• The developed machine learning model accurately predicts RSV infections in hospitalized children.
• This tool offers a rapid and cost-effective method to aid clinical decision-making.
• Integration into routine pediatric care can enhance patient management and infection control strategies.