Breath Prints for Diagnosing Asthma in Children

Area of Science:

• Biomedical Engineering
• Respiratory Medicine
• Analytical Chemistry

Background:

• Electronic nose (e-nose) technology offers a novel, noninvasive method for detecting volatile organic compounds (VOCs) in exhaled breath.
• Measuring VOCs in breath is a potential tool for identifying airway inflammation, particularly in pediatric asthma.
• The noninvasive nature of e-nose makes it suitable for pediatric applications.

Purpose of the Study:

• To investigate the hypothesis that an electronic nose can differentiate the breath prints of pediatric asthma patients from healthy controls.
• To assess the discriminative ability of e-nose technology in a pediatric population with asthma.

Main Methods:

• A cross-sectional study involving 35 pediatric patients (asthma cases and controls).
• Exhaled breath samples were analyzed using the Cyranose 320 electronic nose device.
• Statistical analyses included principal component analysis (PCA), canonical discriminative analysis (CDA), cross-validation accuracy (CVA), and external validation metrics (accuracy, sensitivity, specificity).

Main Results:

• Internal validation showed cross-validation accuracy (CVA) of 63.63% (Model A) and 90% (Model B) with significant M-distances.
• External validation yielded lower accuracy: 64% (Model A) and 58% (Model B), with sensitivity and specificity around 50-77%.
• No significant differences were found between paired breath sample fingerprints, indicating sample reproducibility.

Conclusions:

• The electronic nose demonstrates potential in discriminating between pediatric asthma patients and controls based on breath volatile organic compounds.
• The discrepancy between internal and external validation accuracy highlights the need for further refinement and validation of e-nose technology for clinical application in pediatric asthma diagnosis.