Cardiac imagers need to understand AI as it enters clinical use and ACC guidelines

The academic discussions on AI and its future applications at conferences are largely over. What is now being discussed are actual U.S. Food and Drug Administration (FDA)-cleared products sold on the market and in clinical use.

According to Nicol, AI has now become real, and the first cardiology AI algorithm is now included in practice guidelines in both Europe and the United States. The ACC/AHA 2021 Chest Pain Guidelines include the recommended use of AI-driven fractional flow reserve hemodynamic flow measurements derived from noninvasive CT imaging (FFR-CT). 

"If you had said to me 15 years ago we were going to have some sort of computational fluid dynamics tool in the U.S. chest pain guidelines, you would have been laughed at. Everyone would have said you were crazy. But that is the reality, we see FFR-CT in the international guidelines, based on evidence. That will be the first, I suspect, of many," Nicol said. 

[AI-powered CCTA outperforms usual care when evaluating stable chest pain]

Leaders leaving clinical posts to head AI companies

The seriousness of the AI trend can be highlighted by the fact that two former SCCT presidents and key opinion leaders in cardiac CT recently left their academic and clinical positions to head new AI companies. These leaders believe AI will be a paradigm shift in how cardiac patients are treated in the next decade. Most notable are SCCT past-presidents Todd Villines and James Min, who both entered the industry in the past couple years to push AI technologies they believe can significantly move the needle on patient care. These moves caught the attention of many cardiac CT imagers.

"We have seen a number of well-respected academics over the past few years moving across to become the CMOs or CEOs of these kinds of AI companies," Nicol said. "These are people whose inquisitive, academic personas have not changed, it's just that they are now working for industry. Maybe they just felt they could move things more rapidly in a non-academic world, but they are still passionate, driven people who want to see event rates in cardiovascular patients drop."

Cardiologists and radiologists need to understand how AI works to challenge it

Nicol chaired an all-day pre-conference seminar on the use of AI in cardiac imaging at SCCT 2022 and spoke to a full room. He said it is clear clinicians want more information on what AI can do and how it may help with efficiency and boosting clinical outcomes. Medical societies and clinicians need to be directing and nudging industry partners toward practical AI solutions that solve real problems in medicine, or to tweak AI with real-world input from clinicians, Nicol said. 

"We the clinicians really need to own this," Nicol explained. "And you can really only own this and challenge it if you understand it and the strengths and weaknesses of AI. We are not trying to change the whole radiology/cardiology community into programmers, but they need to understand how those programs work, even if they do not understand all the strings of computer code."

For this reason, he said SCCT will continue to have an AI track at its conferences into the future. The Radiological Society of North America (RSNA) has also embraced AI with its own AI track at its massive annual meetings. This includes a dedicated expo floor area to house the large number of AI vendors that now exhibit at RSNA.

FDA has now cleared more than 500 AI algorithms, mostly for medical imaging 

While AI discussion or vendors can be found at most medical conferences today, AI has taken a very high-profile, front seat at RSNA, SCCT, the America Society of Echocardiography (ASE) and other imaging conferences because most of the commercialized AI applications are for medical imaging. 

As of January 2023, there were more than 520 FDA-cleared AI medical algorithms available in the United States. Radiology, with 396, has the vast majority of AI models. Cardiology is second with 58, although a sizable number of the radiology AI algorithms are specific to cardiac imaging. A very distant third place for AI algorithms is the specialty of hematology, with 14. 

While the first AI app was cleared by the FDA in 1995, and the majority, more than 300 algorithms, have been cleared by the FDA in just the past four years. That number is also accelerating. At the most recent SCCT and the American Heart Association (AHA) meetings, at least 10 new AI algorithms for cardiology were shown for the first time as works-in-progress, or just having been cleared but not yet on the FDA's published list.

Read more in the article FDA has now cleared more than 500 healthcare AI algorithms.

See the complete list of FDA-cleared algorithms here