Machine Learning-Assisted Diagnosis Model for Chronic Obstructive Pulmonary Disease

Chronic obstructive pulmonary disease (COPD) is a long-term, irreversible, and progressive respiratory that often leads to lung function decline. Pulmonary tests (PFTs) provide valuable information for diagnosing COPD; however, they are underutilised in clinical practice, with only subset of test values being used decision making. The final diagnosis requires combining PFT results patient information, symptoms, other tests, such as imaging blood analysis. This study aims comprehensively utilise all the testing PFTs assist COPD. Various machine learning models, logistic regression, support vector (SVM), k-nearest neighbour (KNN), random forest, tree, XGBoost, have been employed establish COPD assistance models. XGBoost model, trained features extracted by group LASSO algorithm, achieved best performance, an area under receiver operating characteristic curve (ROC) 0.90, 88.6% accuracy, 98.5% sensitivity. model can doctors early prediction