A Novel Machine Learning Approach for Prediction of Chronic Obstructive Pulmonary Disease

Chronic Obstructive Pulmonary Disease (COPD) is the most serious chronic disease that begins slowly and progresses to eventual lung cancer. COPD diseases must be treated early before they become serious. Our goal is to predict COPD on the basis of extracted features from region of interest of Computed Tomography (CT) images of patients. We proposed machine learning model using supervised machine learning classifiers to predict COPD. In this paper we discussed prediction of CODP using machine learning approach. The early identification and prediction of lung diseases have become a necessity in the research, as it can facilitate the subsequent clinical management of patients. The proposed prediction models predict COPD and healthy (Non-COPD) efficiently from standard derived features set from CT images of COPD machine learning dataset. Our model used derived features set and trained model using machine learning classifier are Stochastic Gradient Descent, Logistic Regression, Multilayer Perceptron, Random Forest and XG boost applied with optimal parameter selection using distinctive approach which improves the performance of proposed Machine learning classifier. Overall scenario is novel approach for the prediction of COPD using proposed supervised machine learning algorithm.