Artificial intelligence-based machine learning protocols enable quicker assessment of aortic biomechanics: A case...

Area of Science:

• Biomedical Engineering
• Cardiovascular Imaging
• Artificial Intelligence in Medicine

Background:

• Analyzing abdominal aortic aneurysm (AAA) biomechanical wall stresses is complex and time-consuming.
• Clinical adoption of image-based biomechanical analysis is limited by expertise and time requirements.
• Longitudinal tracking and advanced analysis are crucial for managing AAAs.

Purpose of the Study:

• To demonstrate the feasibility of expediting advanced aortic image analysis for a single patient longitudinally.
• To showcase the utility of an artificial intelligence (AI)-based classifier for predicting patient outcomes.
• To present the workflow for analyzing a juxtarenal aortic aneurysm case.

Main Methods:

• Utilized a multidisciplinary, multi-institute team approach.
• Employed advanced image analysis protocols, expedited for a single patient.
• Applied a previously trained AI classifier for outcome prediction.

Main Results:

• Successfully expedited advanced aortic image analysis on a longitudinally tracked patient.
• Demonstrated the AI classifier's accuracy in predicting patient outcomes.
• Detailed the workflow for a specific case of juxtarenal aortic aneurysm management.

Conclusions:

• Expedited advanced aortic image analysis is feasible for longitudinal patient tracking.
• AI-based outcome prediction can serve as a viable alternative to serial surveillance for AAAs.
• The described workflow provides a streamlined approach to managing complex aortic aneurysms.