Forecast the Exacerbation in Patients of Chronic Obstructive Pulmonary Disease with Clinical Indicators Using Machine...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Pulmonology

Background:

• Assessing and predicting the severity of chronic obstructive pulmonary disease (COPD) exacerbations during hospitalization is critical, yet current methods are inadequate for acute phases.
• Traditional prognostic methods for COPD exacerbations are time-consuming, necessitating advanced solutions for timely patient management.

Purpose of the Study:

• To develop and evaluate an AI-driven system for accurately identifying the severity of COPD exacerbations.
• To enhance the monitoring and treatment strategies for patients experiencing acute COPD exacerbations through improved prediction.

Main Methods:

• A soft voting ensemble (SVE) classifier was proposed, integrating five machine learning models: Random Forests (RF), Support Vector Machine (SVM), Gradient Boosting Machine (GBM), XGBoost (XGB), and K-Nearest Neighbors (KNN).
• The ensemble model was trained and validated using a dataset comprising 24 distinct features relevant to COPD patient severity.

Main Results:

• The SVE classifier achieved high performance metrics, including 91.08% accuracy, 90.77% precision, 91.36% recall, 91.07% F-measure, and an AUC score of 96.87%.
• The proposed SVE model demonstrated superior performance compared to individual machine learning classifiers for COPD severity prediction.

Conclusions:

• The SVE classifier with 24 features offers a robust and accurate method for early estimation of COPD patient severity.
• This AI approach can significantly assist respiratory physicians in guiding treatment strategies and improving the prognosis for COPD patients.