Conditions like aortic aneurysms (AA) require timely diagnosis and meticulous management to prevent life-threatening complications. Recent advancements in AI have revolutionized the way healthcare professionals approach AA care – both in the domain of thoracic aortic aneurysm (TAA) and abdominal aortic aneurysm (AAA) – leading to significant improvements in patient outcomes.
In this post, we’ll explore the benefits of AI-enhanced AAA patient care at Yale New-Haven Health, its impact on diagnosis and disease awareness and the implementation process within health systems. Our insights here are drawn from a webinar featuring Edouard Aboian, MD, Vascular and Endovascular Surgeon and Uwe Fischer, MD, Vascular and Endovascular Surgeon, both of Yale New-Haven Health.
1. The Benefits of AI-Enhanced AAA Patient Care
There is a great value of AI to put AAA patients on the radar of physicians, but its benefits don’t stop there. What about patients who entered a health system for an unrelated, acute reason, and their AAA was incidentally found on a CT conducted for other reasons? This is where AI has proven to be a true game changer: in the finding and management of AAA. The primary benefits include:
- Early and Accurate Flagging: Whether on imaging or analyzing radiology reports, AI can identify enlarged areas of the aorta and flag potential aneurysms with high precision. In the case of AAA care, early detection is crucial for initiating a surveillance plan or for timely intervention if warranted.
- Automated Alerts and Follow-Ups: AI systems can automatically alert healthcare providers about critical findings (with site-specific user-set thresholds), expediting treatment for acute patients and ensuring that subacute patients are not lost to follow-up. This enhances clinical outcomes and follow-up adherence and reduces the risk of missed diagnoses.
- Efficient Workflow Integration: AI seamlessly integrates into existing workflows, allowing for faster decision-making and streamlined patient management processes.
2. Impact on AAA Disease Awareness and Management
Dr. Fischer and Dr. Aboian shared valuable insights on AI’s impact on the AAA diagnosis process at Yale, as well as overall disease awareness and patient management.
AI-Assisted AAA Management
Before the implementation of AI, the pathway for managing AAA patients involved several steps, often leading to delays and missed follow-ups. A radiologist would identify an aneurysm on a CT scan, prompting a manual notification to the ordering physician. However, many patients were lost to follow-up due to this cumbersome process.
With AI, the system now automatically evaluates CT scans for enlarged abdominal aortas. If an aneurysm is confirmed, an alert is sent to the healthcare provider, prompting immediate action. This streamlined process ensures that patients are swiftly referred to vascular surgery for further evaluation and management.
The implementation of AI led to significant improvements in AAA treatment at Yale, as noted in two research studies:
The first study captured their initial experience which found that, during a 5 month period:
- 446 aortic aneurysm patients were identified, with 188 patients having aneurysms larger than 4cm.
- 24% of these aneurysms >4cm, were newly diagnosed, highlighting AI’s role in uncovering previously undetected cases.
The second study focused on the utilization of AI for initial evaluation and follow-up of AAA patients (and compared the results to the pre-AI time period):
- The number of relevant patients receiving follow-up consultation within 6 months of index imaging increased from 18% to 42% post-implementation
- Mean time from index imaging to clinic evaluation decreased from 83 days to 22 days
- Patients with long term follow-up scheduled increased from 65% to 99%
- Mean time to repair after index imaging decreased from 270 days to 58 days post-AI.
3. Implementation Process of AI at Yale for AAA Care
As outlined by Drs. Fischer and Aboian, AI implementation in a health system involves several critical steps and does not happen overnight:
- Initial Setup and Team Formation: Establishing a dedicated team to oversee the AI integration process is crucial. This team ensured that all necessary protocols and governing structures are in place.
- Algorithm Training and Validation: AI must be trained and validated using historical data to ensure accuracy and reliability.
- Integration with Existing Systems: AI should seamlessly integrate with existing electronic health records (EHR) and imaging systems to allow for efficient data flow and alert generation.
- Continuous Monitoring and Adjustment: Regular monitoring of AI performance and making necessary adjustments ensure optimal functionality and accuracy.
The Path Forward with AI for AA Care
The integration of AI into AA care marks a significant advancement in care delivery. By enhancing disease awareness, streamlining patient management and ensuring timely intervention, AI has the potential to improve patient outcomes and even save lives. As Drs. Fischer and Aboian highlighted, the journey of AI is just beginning, and its impact on AAA at Yale is a testament to the power of technology transforming patient care.
For health systems considering the implementation of AI, the benefits are clear:
- A more streamlined approach to managing complex conditions
- Improved follow-up management and patient retention, and
- Better patient outcomes
By embracing AI, we can pave the way for a future where advanced technology improves patient capture and ultimately, their well being.
Click here to watch the full webinar, “Transforming Aortic Aneurysm Patient Care“
