"What is really challenging the cardiac imaging community right now is the rise of AI," Nicol said. "For the clinician, it is that balance of embracing the new technology, but just as we need to judge clinical evidence, we are now having to learn how to judge new technologies, and it's not easy because it is not what we are used to."

He said clinicians are now having to learn what questions to ask and how to evaluate AI software and technical information that is outside of their usual scope. But, AI also brings a lot of potential opportunities.

"A few years back at the Radiological Society of North America meeting there was talk about AI replacing the radiologist, but I don't think anyone really believed that. These are decision aid tools to help improve our workflow," Nicol said. "The AI also can bring in big data about the patient, features of their imaging like how many coronary plaques does a patient have, what is the plaque morphology, is their positive remodeling, and what is the overall plaque burden. Putting all that big data into AI allows us to develop models that will give us a new way of looking at things."

He said AI assessment tools are moving cardiology way beyond the transitional look at risk factors for cholesterol, age and smoking status. 

In cardiac imaging, the biggest advance using AI has been commercialized algorithms to offer detailed assessments of the plaque burden rather than just showing there is a narrowing in an artery. This includes automated assessments that detail the types of coronary plaques and the amount of plaque. This may allow for better assessment of risk far beyond calcium scoring. Very precise assessment of plaque volumes also enable the next step in cardiac imaging for a role in coronary disease tracking and prevention. AI can now provide a baseline measurement of plaque from a cardiac CT scan and enable serial assessments over time to show regression of plaques and adjust a patient's risk score from noninvasive imaging alone. This type of AI also can help assess if a patient is noncompliant with statins, or if they need a different class of cholesterol lower drugs to help slow or reverse plaque build up. 

While tools like this appear to have a lot of promise to shift how patients are evaluated and tracked, Nicol said there are questions about the extra cost of these technologies and their real-world value. These conversations will be key, since there is no additional reimbursement for most types of AI, especially in the cardiac CT space. 

"The question is how much value does it add for us? Is it worth the additional in investment?" Nicol said. "Also, which AI do you go for? It is now a lot like buying a car. Do you buy the one that is more sporty and faster, or do you buy the other one that is also sporty, but the seats are terrible." 

Nicol also said clinicians need to make sure they don't buy something presented as "new" that is really just a repackaging of technologies that have been around for years. He said tools that can automatically segment the different types of soft plaques on CT based on Hounsfield units have been around for at least a decade. However, he aded, there is no doubt that the new AI behind most of these algorithms can assess the entire coronary tree in a matter of seconds, where it might take an experienced cardiac imager 30 minutes. 

"Let's take TAVR, for example," he said. "If AI can auto segment the anatomy, take out the aorta and do all the baseline measurements, that might save me 25 minutes on something that I was going to do. But, going back to the coronaries, it is only valuable in saving me 25 minutes if I was going to do it in the first place,"  

Most cardiologists and radiologists may not bother taking the time to meticulously calculate the total plaque burden or segment out the different types of plaque manually. But, if the process is automated and added clinical value, that may change how patients are evaluated. He said there is need for more clinical data from studies to show the added value. 

"What a lot of these AI companies are now saying is that we can use this technology to go back along the clinical pathway, so we are not waiting for people to get ischemia," he said. "This can move us away from primary and secondary prevention to just continuous prevention. So we can stop people from having angina, acute coronary syndrome or a heart attack. I think we do now have the tools and technology to allow us to do that, but I think the really interesting thing will be how you embed these novel technologies into clinical practice." 

In discussions with some of the AI vendors that are focused on plaque, the question was raised as to why a cardiologist would need AI to assess plaque burden numbers to two decimal places. 

"Originally I thought, that makes no sense to me, but that was before speaking with the vendors to understand the context of why theses companies are developing these tools," Nicol said. "But, if you are going to follow people over time, if you are going to use CT as the test to demonstrate if your statin or PCSK9 is working, that makes sense. The patient might be a non-responder to the medication, or they might need a second medication. That's when those kinds of precise quantifications make sense."

He said this is why it is important to know why a company developed an algorithm to do something that appears to be outside of the norm.

There is currently a shortage of cardiologists and radiologists. Can AI help augment that workforce?

"We have an awful lot of technology already that we already have to do a lot of the relatively basic stuff," Nicol said. "We don't pay radiologists or cardiologists to draw lines around things, my 7 and 10 year olds can probably do as good a job on that. We also have AI systems that can do a first read and determine the first 10 cardiac CT scans are normal, and these 12 are abnormal. So you can get a cup of coffee and blast through the normals and then determine when you are fresh, or in the morning, go through the difficult ones. These technologies exist already, but we are not leveraging them. What we are really pay a radiologist or cardiologist for is to put the findings into context."