Actionable AI Models for Predicting Patient Risk in Chronic Obstructive Pulmonary Disease

Watch a technical conference talk exploring the development and deployment of AI risk prediction models for Chronic Obstructive Pulmonary Disease (COPD) patients in the UK healthcare system. Learn how Lenus Health created and implemented the first live AI risk prediction model, focusing on 12-month mortality risk and 90-day hospital readmission predictions. Discover the process of working with NHS clinicians to design practical applications for identifying high-risk patients using data from over 28,000 patients' electronic health records. Examine the models' performance metrics, including ROC-AUC scores of 0.85 and 0.73 for mortality and readmission predictions respectively, and understand how SHAP explainability techniques provide both global patterns and patient-specific insights. Gain valuable insights into real-world challenges of implementing AI in clinical settings, including scalability considerations, continuous monitoring requirements, regulatory compliance, and the complexities of integrating artificial intelligence into existing clinical workflows.