Research protocol for the diagnostic accuracy of intelligent portable impulse oscillometry across different respiratory diseases: a multicentre, cross-sectional study

Introduction: Early clinical manifestations of chronic respiratory diseases are often non-specific, and substantial symptom overlap exists among different diseases, making early differential diagnosis challenging. Conventional spirometry requires a high degree of patient cooperation, which limits its applicability in primary care settings and special populations. Impulse oscillometry (IOS) reflects respiratory mechanical characteristics during tidal breathing and offers several advantages, including convenience, non-invasiveness and low dependence on patient cooperation. This study aims to develop a multi-disease intelligent diagnostic model based on a portable IOS device and to systematically evaluate its diagnostic performance across different respiratory diseases.

Methods and analysis: This is a multicentre, cross-sectional, observational diagnostic accuracy study. The study population includes healthy controls and patients with chronic obstructive pulmonary disease, bronchial asthma, fibrotic interstitial lung disease, bronchiectasis and central airway obstruction. After completing the required medication washout period, all participants will undergo IOS, fractional exhaled nitric oxide measurement, exhaled carbon dioxide monitoring, cough sound collection and spirometry during a single study visit. Depending on the disease type, participants will also undergo bronchodilator testing, bronchial provocation testing or diffusing capacity testing. All index tests and reference standard assessments will be performed at a single visit; no longitudinal follow-up measurements are planned. A multiclass disease diagnostic model will be developed based on IOS parameters. The primary outcome is the diagnostic performance of the model across different respiratory diseases, which will be evaluated in terms of discrimination and calibration.