FDA clears software to reduce ‘guesstimation’ when assessing atherosclerotic risk
Click Here to Manage Email Alerts
Key takeaways:
- The FDA cleared a semiautomated software to aid in plaque visualization and classification.
- The AI-assisted software is designed to generate 3D models of plaque composition to improve ASCVD risk stratification.
Elucid announced it received 510(k) clearance from the FDA for its noninvasive imaging analysis software to help quantify and classify plaque morphology and diagnose CVD.
The software (PlaqueIQ) is designed to utilize first-line diagnostic coronary CT angiography to produce clinically meaningful interactive reports based on the gold standard, histology ground-truth to help visualize the plaque, according to an industry press release.
“Currently, for prevention of MI, we depend on patients' presentation of symptoms and risk factor analysis, which means that typically, if a person does not have risk factors or their calculated atherosclerotic cardiovascular disease (ASCVD) risk score is low, and they are asymptomatic they are not referred to a cardiologist. Even if they are referred, they are often reassured without any further workup,” said Amir Ahmadi, MD, FACC, FRCPC, FSCCT, clinical associate professor of medicine and cardiology at Icahn School of Medicine at Mount Sinai and co-director of the cardiac ICU at Mount Sinai Fuster Heart Hospital at Morningside, in an interview with Healio. “However, when you look at 1 million people who have had a heart attack and look back, you see that an incredibly large number of them were in the low- to intermediate-risk group and about two thirds of them did not have symptoms prior to their MI . If you had seen them a week prior to their MI as a cardiologist, you would have been unable to identify them to be at risk or prescribe statins by our current methods.”
“Now we have the technology to move away from merely 'guesstimating' the risk of disease to actually visualizing the disease,” he said. “Identifying and quantifying plaque at the vessel level, and being able to assess the risk of each individual lesion to predict who is truly at risk—independently of risk factors and the status of symptoms—is crucial for changing our understanding, improving risk stratification, and enhancing patient outcomes.”
The software is developed to quantify and characterize plaque and its components — such as lipid-rich necrotic core — and may enable earlier identification of high-risk plaque before symptoms or major adverse events occur, according to the release.
“PlaqueIQ is a reliable tool. The algorithms are trained by histology and can detect plaque components in a very reliable fashion. It can quantify and tell the morphology of the plaque and we can compare from time point A to time point B, whether the patient is responding to treatment,” Ahmadi told Healio.
Physician users of the software would send patient images to the company with a single mouse click, where image-restoration algorithms to reduce motion and calcium blooming artifacts and trained analysts create a 3D model of the patient’s coronary arteries, according to the release.
“In the current version, the image is sent to Elucid, and it’s a semiautomated process from there. The technicians will do the image processing. After they do the initial segmentation of the coronary tree, the computer processes it and the report is autogenerated and ready to send back to the physician,” Ahmadi said. “Currently, this in under 24-hour turnaround time, but as with new technology, over time this turnaround time is going to be significantly shortened.”
The software is currently undergoing beta testing. The company stated it anticipates a limited release of the software in the fourth quarter of 2024.
“At the lesion level, we need to be able to risk stratify which lesions are going to have an increased change of rupture and cause a heart attack, and that depends on lesion level morphology, which very much potentially depends on a necrotic core and its precise volumetric assessment ,” Ahmadi told Healio. “If shown in a prospective fashion that [PlaqueIQ] is predictive of an event, it can help us guide management.”
“What Elucid did with PlaqueIQ is define something called a noncalcified matrix as opposed to noncalcified plaque. The reason for that is that there is always a gray zone to completely, accurately differentiate between plaque and the wall. That gray zone could very well be the source of error in measurements in the same patient over time,” Ahmadi said. “By defining the noncalcified matrix, that differentiation between time point A and B is going to be much more accurate and related to the actual change in plaque volume over time in response to treatment.”
The company stated it is also pursuing a software indication for noninvasive measurement of fractional flow reserve, according to the release.
For more information:
Amir Ahmadi, MD, FACC, FRCPC, FSCCT, can be reached at 440 W. 114th St., 2nd Floor, Suite 220, New York, NY 10025.
Collapse
Click Here to Manage Email Alerts