Related Experiment Video
Updated: Aug 6, 2026
09:01
An In vitro Model to Study Immune Responses of Human Peripheral Blood Mononuclear Cells to Human Respiratory Syncytial Virus Infection
Published on: December 10, 2013
8.0K
Hemogram-based decision tree models for discriminating COVID-19 from RSV in infants
Dejan Dobrijević1,2, Ljiljana Andrijević1, Jelena Antić1,2
1Faculty of Medicine, University of Novi Sad, Novi Sad, Serbia.
Journal of Clinical Laboratory Analysis
|March 27, 2023
Summary
Decision tree models can rapidly distinguish between COVID-19 and RSV infections in infants. The optimized forest model demonstrated superior performance, aiding in timely clinical decisions.
Area of Science:
- Medical Informatics
- Pediatric Infectious Diseases
- Machine Learning in Healthcare
Background:
- The current pandemic highlights the need for rapid diagnostic tools.
- Distinguishing between COVID-19 and RSV in infants is clinically significant.
- Decision tree algorithms offer efficient and reliable decision-making capabilities.
Purpose of the Study:
- To develop and evaluate decision tree algorithms for differentiating COVID-19 from RSV in infants.
- To assess the performance of various decision tree models in this diagnostic task.
Main Methods:
- A cross-sectional study involving 77 infants (33 with SARS-CoV-2, 44 with RSV).
- Utilized 23 hemogram-based features to construct decision tree models.
- Employed a 10-fold cross-validation method for model evaluation.
Main Results:
- The Random Forest model achieved the highest accuracy at 81.8%.
- The optimized forest model showed superior sensitivity (72.7%), specificity (88.6%), PPV (82.8%), and NPV (81.3%).
Conclusions:
- Random Forest and optimized forest models show potential for rapid clinical decision-making.
- These models can aid in suspected SARS-CoV-2 and RSV cases before definitive testing.
- Accelerated diagnosis can improve patient management in pediatric respiratory infections.
