Cardiac CT's future could include AI-driven prevention, noncontrast imaging
Cardiac computed tomography (CT) could expand well beyond coronary artery assessments as artificial intelligence (AI), radiomics and new scanner technologies enable clinicians to identify high-risk patients through the use of low-cost, noncontrast scans.
Ed Nicol, MD, head of cardiovascular CT at Royal Brompton Hospital in London and a former president of the Society of Cardiovascular Computed Tomography (SCCT), discussed emerging noncoronary applications of cardiac CT at SCCT 2026. In the above video interview with Cardiovascular Business, he highlighted three major areas of development: structural heart disease, congenital heart disease and the use of cardiac adipose tissue as a source of predictive information.
In structural heart disease, he said AI could initially have a major impact by improving efficiency rather than attempting to fully automate diagnosis. Segmentation and other workflow tools already exist, but have not yet penetrated clinical practice as extensively as expected. This is especially important as the volume of cardiac CT scans continues to rise.
“The workforce is not really catching up at the same kind of speed,” Nicol said.
But, he added, AI tools that reduce the amount of human labor required for image processing could provide value, even if they are not directly reimbursed.
Nicol expects AI to eventually play a greater role in predicting future cardiovascular events, although he questioned whether fully autonomous systems will become widespread because clinicians will likely retain legal responsibility for all decisions.
Congenital heart disease is another potential growth area. As more children with congenital heart disease survive into adulthood, cardiologists increasingly encounter patients with complex anatomy and prior surgical repairs. CT can help determine what procedures patients previously underwent and evaluate whether new symptoms are related to those interventions.
Newer technologies such as photon-counting CT also could increase the role of CT in congenital heart disease. He said the technology promises to deliver more detailed imaging and help lower radiation doses.
Inflammation imaging of the heart may predict heart failure and AFib
Nicol said a very promising area of development is the analysis of fat surrounding the heart using fat attenuation index (FAI) imaging to visualize inflammation. Research has shown coronary fat inflammation can predict myocardial infarction (MI) risk, and the same may be true of for epicardial fat layers in predicting other types of cardiomyopathies. He pointed to emerging data suggesting that analysis of adipose tissue around the ventricles may help predict future heart failure, while assessment of tissue around the left atrium could help identify patients at risk of developing atrial fibrillation or stroke.
The ability to perform some of these assessments using non-contrast CT also could have major implications for preventive cardiology screenings in the future.
“If you could do MI prediction from the coronaries, you can do heart failure prediction from the ventricular tissue or the adipose tissue around the ventricles, and then you can do a stroke risk. Then all of a sudden a relatively low-cost, non-contrast study can be a predictive tool in preventative cardiology,” Nicol said.
Such an approach could eventually shift cardiac CT from identifying established disease toward identifying risk before symptoms or significant plaque develop.
Nicol acknowledged that broad screening of asymptomatic populations would be controversial, but said the potential societal and economic benefits warrant investigation. He suggested that future research could even explore whether routine non-contrast cardiac CT screening might be appropriate for certain age groups.
The challenge will be determining what clinicians should do with the additional risk information. Nicol said imaging must ultimately be linked to effective preventive interventions, including medications and lifestyle changes. The information one how these tests work also need to be translated easily into plain language a patient can understand.
With large datasets involving tens or hundreds of thousands of patients now being analyzed, Nicol expects evidence supporting these predictive applications to develop rapidly.
“The field is moving incredibly fast,” he said. “What seemed a distant dream 18 months, two years ago is now emerging as a reality.
Editor's note: Nicol is a member of the scientific advisory board for Caristo, which is developing this FAI technology.
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: [email protected]