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

Background and objective Spirometry, which is the most commonly used technique for asthma diagnosis, is often unsuitable for small children as it requires them to follow exact instructions and perform extreme inspiration and expiration maneuvers. In contrast, impulse oscillometry (IOS) is a child-friendly technique that could serve as an alternative pulmonary function test (PFT) for asthma diagnosis and control in children as it offers several advantages over spirometry. However, the complex test results of IOS may be difficult to be understood by practitioners due to its reliance on mechanical and electrical models of the human pulmonary system. Recognizing this reality, computer-aided decision systems could help to improve the utility of IOS. The main objective of this paper is to understand the current computer-aided classification research works on this topic. Methods This paper presents a methodological review of research works related to the computer-aided classification of peripheral airway obstruction using the IOS technique, which is focused on, but not limited to, asthmatic children. Publications that focused on computer-aided classification of asthma, peripheral dysfunction and/or small airway impairment (SAI) based on impulse oscillometric features were selected for this review. Results Out of the 34 articles that were identified using the selected scientific web databases and topic-related parameters, only eight met the eligibility criteria. The most relevant results of the articles reviewed are related to the performance of the different classifiers using static features which are solely based on the first pulmonary function testing measurements (IOS and spirometry). These results included an overall classifiers' accuracy performance ranging from 42.24% to 98.61%. Conclusion There is still a great opportunity to improve the utility of IOS by developing more computer-aided robust classifiers, specifically for the asthmatic children population as the classification studies performed to date (1) are limited in number, (2) include features derived from tests that are not optimally suitable for children, (3) are solely bi-class (mostly asthma and non-asthma) and therefore fail to include different degrees of peripheral obstruction for disease prevention and control and (4) lack of validation in cases that focus on multi-class classification of the different degrees of peripheral airway obstruction.

Analysis of impulse oscillometric measures of lung function and respiratory system model parameters in small airway-impaired and healthy children over a 2-year period.

BACKGROUND: Is Impulse Oscillometry System (IOS) a valuable tool to measure respiratory system function in Children? Asthma (A) is the most prevalent chronic respiratory disease in children. Therefore, early and accurate assessment of respiratory function is of tremendous clinical interest in diagnosis, monitoring and treatment of respiratory conditions in this subpopulation. IOS has been successfully used to measure lung function in children with a high degree of sensitivity and specificity to small airway impairments (SAI) and asthma. IOS measures of airway function and equivalent electrical circuit models of the human respiratory system have been developed to quantify the severity of these conditions. Previously, we have evaluated several known respiratory models based on the Mead's model and more parsimonious versions based on fitting IOS data known as extended RIC (eRIC) and augmented RIC (aRIC) models have emerged, which offer advantages over earlier models. METHODS: IOS data from twenty-six children were collected and compared during pre-bronchodilation (pre-B) and post- bronchodilation (post-B) conditions over a period of 2 years. RESULTS AND DISCUSSION: Are the IOS and model parameters capable of differentiating between healthy children and children with respiratory system distress? Children were classified into two main categories: Healthy (H) and Small Airway-Impaired (SAI). The IOS measures and respiratory model parameters analyzed differed consistently between H and SAI children. SAI children showed smaller trend of "growth" and larger trend of bronchodilator responses than H children.The two model parameters: peripheral compliance (Cp) and peripheral resistance (Rp) tracked IOS indices of small airway function well. Cp was a more sensitive index than Rp. Both eRIC and aRIC Cps and the IOS Reactance Area, AX, (also known as the "Goldman Triangle") showed good correlations. CONCLUSIONS: What are the most useful IOS and model parameters? In this work we demonstrate that IOS parameters such as resistance at 5 Hz (R5), frequency-dependence of resistance (fdR: R5-R20), reactance area (AX), and parameter estimates of respiratory system such as Cp and Rp provide sensitive indicators of lung function and have the capacity to differentiate between obstructed and non-obstructed airway conditions. They are also capable of demonstrating airway growth-related changes over a two-year period. We conclude that the IOS parameters AX and the eRIC model derived parameter Cp are the most reliable parameters to track lung function in children before and after bronchodilator and over a time period (2 years). Which model is more suitable for interpreting IOS data? IOS data are equally well-modelled by eRIC and aRIC models, based on the close correlations of their corresponding parameters - excluding upper airway shunt compliance. The eRIC model is a more parsimonious and equally powerful model in capturing the differences in IOS indices between SAI and H children. Therefore, it may be considered a clinically-preferred model of lung function.

Electrical circuit models of the human respiratory system reflect small airway impairment measured by impulse oscillation (IOS).

The use of the forced oscillatory input impedance parameter, frequency-dependence of Resistance (fdR), to assess small airway impairment (SAI) has not been widely accepted due to concern about the effects of "upper airway shunt" on oscillometric resistance and low frequency reactance. On the other hand, recent medical studies suggest that low frequency reactance is a very sensitive index of treatment intervention directed at small airways. The present study was undertaken to analyze and compare Impulse Oscillometry (IOS) resistance and reactance data with model-derived indices of small airway function from two models of the respiratory impedance, one with, and the other without an element for upper airway shunt capacitance. Fifty six patients with stable chronic obstructive lung disease of varying severity due to Cystic Fibrosis (CF) and 21 patients with asthma were evaluated by IOS testing. IOS data were input into the augmented RIC (aRIC) model with an upper airway shunt capacitance, and the extended RIC (eRIC) model, without a shunt capacitance element. Model-derived indices were compared between the two models for CF patients separately from asthma patients. We conclude that IOS indices of SAI are modeled equally well with or without upper airway shunt capacitance, and do not seem to be dependent on upper airway shunt capacitance.

Impulse oscillometric features of lung function: towards computer-aided classification of respiratory diseases in children.

Asthma is the most prevalent chronic respiratory disease in children. Reliable and patient-friendly instruments and methods are required to help pulmonologists accurately detect asthma with acceptable clinical accuracy, specificity and sensitivity. Impulse Oscillometry System (IOS) based on the Forced Oscillation Technique (FOT) has been successfully used to measure lung function in children with a high degree of sensitivity and specificity to small airway dysfunction (SAD). IOS measures the mechanical impedance of the respiratory system. Equivalent electrical circuit models of lung function have been developed that can be used to quantify severity of SAD. It has been shown that impulse oscillometric parameters as well as parameter estimates of these electrical models provide useful indicators of lung function and therefore have the potential to be used as sensitive features for computer-aided classification of pulmonary function in health and disease.