Use of AI opportunistic screening in CT for cardiovascular disease
A promising area of artificial intelligence (AI) in radiology is its use to opportunistically screen patients for various diseases on any computed tomography (CT) scans they happen to receive.
Researchers at Brigham and Women's University recently conducted a study using AI to screen any CT scans signs of heart disease in patients with elevated levels of inflammation already involved in their rheumatology clinic. The AI screening of more than 2,000 patients found over 50% had detectable coronary calcium levels showing they already have early coronary artery disease (CAD).
Many patients undergo chest CT scans for conditions like pneumonia or lung cancer, completely unrelated to cardiac disease. However, these scans can also reveal coronary calcifications as incidental findings, a key indicator of CAD, but the calcium often goes unreported or is not followed up on. AI now offers an opportunity to look at scans, even retrospectively, to find patients at higher coronary disease risk and flag them for followup long before they have any symptoms.
“We’re finding that when you apply AI to these non-cardiac chest CTs, it’s possible to detect coronary calcifications in patients who may otherwise go unnoticed," explained Ron Blankstein, MD, FACC, FASNC, MSCCT, FASPC, associate director, cardiovascular imaging program, director, cardiac computed tomography, Brigham and Women’s Hospital, who was involved in the study.
Blankstein explained that AI can now analyze these images and quantify the calcified plaque present in the coronary arteries, even if the scan was intended for a different purpose.
Worked with a team led by Brittany Nicole Weber, MD, PhD, cardiac imager and director of the Cardio-Rheumatology Clinic at Brigham and Women’s Hospital, in collaboration with Nanox AI, they presented their findings at the Society of Cardiovascular Computed Tomography (SCCT) 2024 meeting. The AI was used to analyze CT scans of about 2,500 patients with autoimmune diseases, such as lupus and rheumatoid arthritis. The study showed more than 50% of these patients has with higher levels of calcified plaque had significantly increased risks of cardiovascular events like heart attacks. Inflammation on coronary plaques is one of the key ingredients that causes heart attacks, and the study looked at whether these patients with inflammatory diseases were at a high risk, and the answer was yes.
This opportunistic screening is particularly relevant in patients with inflammatory conditions, where inflammation accelerates plaque formation and increases the likelihood of events like plaque rupture, leading to heart attacks. Blankstein emphasized that inflammation is a potent risk factor for heart disease, and these findings provide valuable information to physicians, enabling them to adopt more aggressive prevention strategies for at-risk patients before they become symptomatic, or presented to an emergency room with a heart attack.
The study's outcomes also contribute to the growing body of evidence that inflammation plays a crucial role in cardiovascular health, making AI-powered screening a powerful tool in early intervention. “Once we detect calcifications in these patients, we can intensify their treatment, whether it’s lowering LDL cholesterol, managing blood pressure, or addressing inflammation with medications like colchicine," Blankstein explained.
The future of AI in cardiovascular imaging holds even greater promise. Blankstein envisions AI seamlessly integrating into electronic medical records (EMR) systems, flagging high-risk patients who may have been overlooked. Numerous cardiology experts believe CT screenings will become a primary prevention tool to screen patients for low levels of disease so they can begin drug treatments years or decades before they every would become symptomatic or suffer a heart attack. AI opportunistic screenings would go a long way to finding patients who would benefit from early statin therapy to avoid any cardiac issues later in life.
“The potential of AI to screen millions of chest CTs stored in hospital PACS systems and identify patients with hidden coronary artery disease could revolutionize preventive cardiology,” Blankstein said. This technology could provide a significant boost to population health screening efforts by moving beyond traditional risk factors and looking directly at diagnostic imaging results.
As AI continues to enhance the accuracy and efficiency of cardiovascular disease detection, Blankstein believes it will lead to earlier interventions and better outcomes for patients. "It’s not just about identifying the disease, but also improving patient and physician behavior to mitigate the risks,” he said.
This growing interest in using AI for opportunistic screening may soon lead to its widespread adoption, positioning it as a valuable tool in reducing the burden of cardiovascular disease globally. Opportunistic screens also hold an economic value for health systems, where they can screen patients without devoting additional resources for dedicated screening programs and taking up additional scanner time. While there is no reimbursement at this time for opportunistic AI screening, health systems can gain ROI based on the additional patients they can bring into cardiology programs and the downstream testing, therapy and office visits downstream.
Read and hear more about this study in a interview with lead author Brittany Nicole Weber, MD, PhD, Brigham and Women’s Hospital, in the video Opportunistic screening with AI could be a game-changer for preventive cardiology.
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: dfornell@innovatehealthcare.com