Abstract: This project aims to predict the necessity of oxygen therapy in patients with Chronic Obstructive Pulmonary Disease (COPD) by leveraging machine learning techniques. The system utilizes various algorithms, including Linear Regression, XGBoost Regressor, Random Forest, and AdaBoost, to process clinical data and predict oxygen saturation levels. With these predictions, healthcare providers can determine whether a COPD patient requires supplemental oxygen therapy. The backend of the system is built using Python, where machine learning models and data processing tasks are carried out to ensure accurate and reliable predictions. On the frontend, the system is designed using HTML, CSS, and JavaScript to create a user-friendly interface that allows healthcare professionals to easily input patient data and view results. By automating the prediction of oxygen needs, this tool can assist in optimizing patient care by offering timely and precise decisions regarding oxygen therapy. This can contribute to more efficient management of healthcare resources while improving patient outcomes, particularly in the management of COPD. The system serves as a decision support tool, helping healthcare professionals provide personalized care based on individual patient requirements.

 Keywords: COPD, oxygen therapy, machine learning, Linear Regression, XGBoost, Random Forest, AdaBoost, oxygen saturation, Python, HTML, CSS, JS.