Computer-aided classification of small airways dysfunction using impulse oscillometric features: a children-focused...

Area of Science:

• Pediatric Pulmonology
• Biomedical Engineering
• Medical Informatics

Background:

• Spirometry, a standard asthma diagnostic tool, is challenging for young children due to complex maneuver requirements.
• Impulse oscillometry (IOS) presents a more accessible, child-friendly pulmonary function test (PFT) for asthma diagnosis and management.
• The complexity of IOS data necessitates computer-aided decision systems to aid practitioner interpretation.

Purpose of the Study:

• To review current research on computer-aided classification methods for interpreting impulse oscillometry data.
• To assess the application of these methods specifically for diagnosing peripheral airway obstruction in children, particularly those with asthma.

Main Methods:

• A methodological review of scientific literature was conducted, focusing on computer-aided classification of peripheral airway obstruction using IOS features.
• Studies involving asthma, peripheral dysfunction, or small airway impairment (SAI) based on IOS data were selected.
• Eligibility criteria were applied to 34 identified articles, with eight meeting the review requirements.

Main Results:

• Reviewed studies primarily focused on static features from initial IOS and spirometry measurements.
• The accuracy of various classifiers ranged widely, from 42.24% to 98.61%.
• Limited research exists, with most studies using bi-class classifications (asthma vs. non-asthma) and not accounting for varying degrees of obstruction.

Conclusions:

• Significant opportunities exist to enhance IOS utility through the development of more robust computer-aided classifiers for pediatric asthma.
• Current classification studies are few, often use non-optimal features for children, lack multi-class discrimination for peripheral airway obstruction severity, and require further validation.