Artificial Intelligence-Based Algorithms Improve Care of Patients with AAA.

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Healthcare
• Vascular Surgery

Background:

• Timely detection and monitoring of abdominal aortic aneurysms (AAAs) are crucial for preventing ruptures and reducing mortality.
• Artificial intelligence (AI) algorithms offer automated detection of AAAs from imaging and radiology reports.
• This study evaluates the impact of AI on AAA detection and care compared to the standard of care.

Purpose of the Study:

• To examine the impact of AI utilization on AAA detection and patient care.
• To compare AI-assisted AAA care with the historical standard of care.
• To assess the influence of AI on the timeline of AAA evaluation and treatment.

Main Methods:

• An AI-based AAA detection and measurement algorithm was deployed, providing notifications for AAAs ≥5 cm.
• A workflow was established for timely referral and evaluation of patients with detected AAAs.
• A pre-AI and post-AI group comparison was conducted, analyzing patient characteristics, evaluation timelines, and follow-up care.

Main Results:

• A greater proportion of patients received initial AAA evaluation post-AI implementation (42% vs. 18%, P < 0.001).
• The post-AI group showed a trend towards a shorter evaluation timeline (22 days vs. 83 days, P = 0.1).
• Long-term follow-up care increased significantly in the post-AI group (45% vs. 30%, P = 0.004), with 99% having scheduled monitoring appointments.

Conclusions:

• AI-assisted protocols significantly increase initial AAA evaluations and long-term follow-up.
• AI implementation correlates with a decreased timeline to initial AAA evaluation.
• AI shortens the time from detection to repair for AAAs measuring ≥5 cm.