How this HCA Midwest cardiology leader uses AI to enhance diagnostics

Imaging cardiologist Vasvi Singh, spoke to Becker’s Hospital Review about the latest way she is using AI to improve patient care. Dr. Singh is director of Overland Park, Kan.-based HCA Midwest Health’s cardiac MRI program and co-director of the cardiac PET and cardiac CT programs. 

Editor’s note: Responses have been lightly edited for clarity and length.

Question: How does AI integration affect your clinical approach to diagnosing and treating coronary artery disease? 

Dr. Vasvi Singh: AI is now being increasingly used in coronary artery disease for risk assessment, for diagnosis and for treatment planning. It offers potential for earlier, simpler and more accurate analysis of patient data using imaging. It has been shown in studies to lead to improved patient outcomes. 

A new therapeutic target that is important to mention in coronary artery disease is coronary inflammation. We know that inflammation is bad, and when it happens in coronary arteries, it leads to cardiac events such as heart attacks, strokes and deaths. Studies have shown that when you treat inflammation, it reduces these cardiovascular events.

The discovery of how to measure inflammation around the coronary arteries using coronary CT imaging was a major breakthrough. If we can find patients who have high inflammation based on imaging, we can offer personalized treatment and therapeutic targets.

Q: Can you describe how this technology integrates into your existing cardiac imaging workflow? 

VS: We measure signal characteristics on coronary CT images using Hounsfield units, or HUs. Caristo’s CaRi-Heart AI technology normalizes HUs to various anatomical, biological, technical and patient-specific factors, and therefore that makes it a very robust indicator of pericoronary inflammation.

At HCA Midwest Health, we are a real-world U.S. clinical site for using this technology in our patients as a part of an important research study. We are aiding efforts toward making it available clinically for our patients who are receiving coronary CT in the future, after FDA clearance, as well. 

Q: How do you envision AI-driven technologies shaping the future of cardiac imaging and personalizing cardiovascular care?

VS: This imaging technology provides a more robust individual risk score in addition to stenosis, plaque and hemodynamic assessment. I think this will become a standard of care when people are referred to assess for coronary artery disease.

AI-driven, CT-based analysis — because it provides patient-specific risk assessment, which is the key for personalized cardiovascular care — will help us identify individuals at high risk of events who may benefit from more aggressive prevention, and then identify individuals who are more likely to benefit from anti-inflammatory therapies. I think in the near future, it will also be used to assess response or lack of response, to therapy or uptitration of therapy. There are absolutely critical uses in a large patient population.