The purpose of this study is to evaluate the accuracy of a novel, fully automatic artificial intelligence (AI)-based 3D tool designed to extract morphological features from pre-operative computed tomography angiography (CTA) images. These features are crucial for the planning of endovascular aneurysm repair (EVAR) procedures in patients suffering from abdominal aortic aneurysms (AAA). Pre-operative analysis of CTA images is a critical component in ensuring the success of EVAR procedures, as it directly influences the planning of stent graft sizing. Inaccurate measurements of aortic-iliac dimensions can lead to incorrect stent graft sizing, which in turn may cause complications during or after the surgery. The ability to automate and standardize these measurements with high accuracy could potentially reduce the risk of such complications, streamlining the planning process.
The study seeks to establish the reliability of the AI tool by comparing its performance with that of manual measurements, traditionally done by experienced vascular surgeons using clinical software. It also explores the time-saving potential of the new AI tool, a key factor in clinical workflows where efficiency is critical. The European Society for Vascular Surgery (ESVS) 2024 guidelines1 for EVAR are used as a reference for the extraction of the relevant morphological data, ensuring that the AI tool adheres to the most up-to-date and standardized practices.