AI is transforming cardiac imaging—often in ways clinicians do not recognize
Artificial intelligence (AI) is quietly transforming nearly every stage of cardiovascular imaging, from scan acquisition to image reconstruction and quantitative analysis. And many clinicians may not even realize they are already using the technology.
Cardiology imaging AI expert Damini Dey, PhD, professor of biomedical sciences, co-associate director of the Biomedical Imaging Research Institute and director of the quantitative image analysis program at Cedars-Sinai, spoke with Cardiovascular Business about this trend. She said AI has matured beyond experimental applications and is now being embedded throughout the cardiac imaging process, often in unseen ways in a seamless clinical workflows. She pointed to polling of attendees in an AI session at ACC.26, where she served as a panelist.
"One of the things that emerged is that people don't know when they're using AI because it's been used behind the surface for so many imaging applications that it's very much integrated and seamless," Dey explained.
The session included audience polling to gauge how clinicians are using AI across imaging modalities and identify barriers to broader adoption, including implementation costs and information technology challenges. Despite rapid advances, the speakers at ACC.26 repeatedly emphasized that AI should enhance, not replace, clinical decision-making done by humans.
"The emphasis is still on clinical workflows and what's best for the patient," she said.
AI already embedded in imaging systems
While attention often focuses on advanced diagnostic algorithms, Dey noted that AI is already improving routine imaging operations long before physicians review the images. Modern CT, MRI and nuclear scanners increasingly use AI to recommend acquisition protocols, assess patients on the table, improve image quality, reduce noise, reconstruct images and optimize low-dose scans automatically.
"You don't know when the scanner suggests the protocol that it's using AI, but it is," Dey explained.
Once images are acquired, AI now performing automated segmentation and quantitative measurements that would otherwise require extensive manual effort by technologists, radiologists or cardiology imagers.
"It increases the efficiency with the caveat that the human has to be in the loop and validate the measurements," she said.
Plaque analysis highlights AI's clinical value
One of the clearest examples of AI's clinical impact is automated coronary plaque analysis. Dey helped develop the FDA-cleared AutoPlaque software, which is now used clinically at Cedars-Sinai to provide detailed quantitative assessments of coronary plaque burden and characteristics from coronary CT angiography (CCTA).
The software allows physicians to evaluate plaque composition and distribution far more comprehensively than traditional visual interpretation alone.
"We find that it can provide additional quantitative analysis and really guide the patient's care," Dey said. "It helps us see the plaque characteristics over the coronary arteries and then try to guide what's right for the patient."
In many cases, the tasks being done by AI would simply not be done if these algorithms did not exist; it is not practical to do these time-consuming tasks so often.
"For plaque analysis, it's definitely not something that could be done without AI. For automation, AI is needed, otherwise it could not be performed," Dey explained.
Clinicians at Cedars-Sinai have embraced the technology because it helps personalize treatment decisions, particularly for preventive cardiology patients.
Reimbursement accelerating AI adoption
The growing availability of reimbursements for advanced cardiac AI is expected to further increase adoption. Recent reimbursement pathways for coronary plaque analysis and CT-derived fractional flow reserve (CT-FFR) have helped justify the investments required to implement these technologies. Dey said reimbursement is important because deploying AI requires a lot of time and money.
Human oversight of AI remains essential
Despite rapid advances in AI, Dey does not foresee autonomous AI replacing physicians anytime soon. While AI can automate measurements and improve workflow, clinicians remain responsible for validating results and making final diagnostic and treatment decisions.
"The human has to be in the loop," she emphasized.
Dey said predicting AI's trajectory over the next five years is difficult, but she expects continued integration across cardiovascular imaging as clinicians become more comfortable with the technology.
"There are growing pains, but in the end you'll have a nirvana of acceptance and you'll see that it makes your life easier and there are huge benefits," she said.
As AI continues moving from standalone software into the core functionality of imaging systems, Dey believes its greatest success may be that users eventually stop noticing it altogether, allowing clinicians to focus less on the technology itself and more on delivering better care for patients.
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]