A support vector machine-based tool for rapid pediatric asthma exacerbation risk assessment: development and nursing...

Area of Science:

• Pediatric Pulmonology
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Childhood asthma presents significant health challenges.
• Traditional asthma assessment methods lack efficiency and accuracy.
• There is a need for improved pediatric asthma risk evaluation.

Purpose of the Study:

• To develop a rapid assessment tool for pediatric asthma exacerbation risk.
• To utilize a support vector machine (SVM) algorithm for risk prediction.
• To evaluate the tool's value in nursing practice.

Main Methods:

• Collected clinical data from pediatric asthma patients.
• Incorporated indicators like eczema, allergic rhinitis, family history, dyspnea, WBC, IgE, and FeNO.
• Developed an SVM risk prediction model and an API for real-time assessment.
• Created the Pediatric Asthma Risk Rapid Tool (PARRT) mini-program.

Main Results:

• The SVM model achieved high accuracy (AUC of 0.9998).
• PARRT implementation significantly reduced patient and report wait times.
• Patient/family satisfaction and nursing staff efficiency showed marked improvement.

Conclusions:

• The PARRT tool demonstrates strong predictive accuracy for pediatric asthma exacerbation.
• PARRT shows significant promise for enhancing clinical utility in pediatric asthma management.
• The tool offers a valuable advancement in nursing practice for asthma care.