Rapid Prediction of Hemodynamic Parameters in Abdominal Aortic Aneurysm Using an AI-Based Surrogate Model

Key Points

• The study aims to explore the relationship between geometric changes and hemodynamic indicators in abdominal aortic aneurysm and to use AI for predicting these parameters.
• Utilized an AI surrogate model called MultiViewUnet to predict hemodynamic parameters.
• Analyzed the geometric characteristics of abdominal aortic aneurysms.
• Compared AI model predictions with computational fluid dynamics results.
• Achieved high accuracy with the AI model, particularly the RRT metric.
• NMAE, NMSE, and NRMSE values approached zero, indicating low prediction error.
• The single inference results closely matched those from traditional computational fluid dynamics.