An Interpretable Feature Selection–Driven Extra Trees Framework for COPD Severity Diagnosis using ABG and Biometric Data

One of the heterogeneous respiratory diseases is
Chronic Obstructive Pulmonary Disease (COPD) that consists of
several symptoms namely breathe shortness, persistent together
with mucus production, chest tightness as well as wheezing,.
Severity assessment in COPD has significant decision making to
guide treatment process but the conventional methods are
majorly highlight on binary diagnosis together with lagging
interpretability to clinical usage. In spite of dataset availability
that consists of patient Arterial Blood Gas along with biometrics,
there are limited research concentrate on systematic integrated
features with understandable Machine Learning (ML) methods
in classifying the severity of the disease towards Global initiative
for chronic Obstructive Lung Disease (GOLD) stages.