Measurements using miniature cameras within the coronary artery can accurately predict whether someone will suffer from a recurrent heart attack. Until now, interpreting these images was extremely complicated and only specialized laboratories could do it. A new study from Radboud University Medical Center shows that AI can ensure that this analysis can be carried over and quickly assess arteries for weaknesses.
A heart attack occurs when a blood clot blocks the coronary artery that supplies blood to the heart. This can occur when atherosclerosis causes arterial stenosis and the heart is produced when there is too little oxygen. Treatment usually involves angioplasty. There, a cardiologist spreads the artery with small balloons, followed by a placement of small tubes, usually called stents. In the Netherlands, this procedure is performed approximately 40,000 times a year.
Predicting recurring events
Nevertheless, about 15% of patients suffering from a heart attack experience another event within two years. To better identify vulnerable spots within the artery that could cause new infarctions, technical physician Joss Tanhauser and Radboudumc physician Rick Volleberg conducted the study together with the team. They analyzed coronary arteries of 438 patients using miniature cameras, developed AI specifically and followed these patients for two years.
This study shows that AI, like the international gold standard, can detect vulnerable spots in the arterial wall and even more accurately predict new infarctions or deaths within two years. What does this mean for the patient? Volleberg explains: “If you have high-risk plaques and know where you are, you can adjust your medication or place a preventive stent in the future.”
Looking inside the artery wall
Miniature cameras use a technique called optical coherence tomography (OCT). Insert through the arm into the bloodstream, capture images of the arteries using near-infrared light, and visualize the vessel wall with microscopic resolution.
“This technique is already used in clinical practice to guide angioplasty and to ensure that the stent is positioned correctly,” explains Thannhauser. 'OCT has been shown to reduce the risk of new infarctions and complications. However, in such cases, the doctor only looks at very small parts of the artery, that is, the site of the infarction. Our research shows that when combined with AI, there is a much greater possibility of mapping the entire ship.
Towards clinical applications using AI
“One of the challenges with this approach is that it is extremely difficult for doctors to interpret OCT images,” says Thannhauser. That's not surprising. Each step generates hundreds of images. Even assessing only the placement of the stent is challenging. Analyzing the entire coronary artery produces a large number of images to manually evaluate. “Currently, only a handful of specialized labs can interpret these images, and we can't see everything. What's more, it's expensive and labor intensive to implement this manually.
That's why the team at Thannhauser has developed an AI that can analyze all images much faster than humans. “AI can already assist doctors during stent placement with OCT,” explains Thannhauser. “AI is one step closer to scanning the entire coronary artery for vulnerable spots in clinical practice. But I hope that it will take years before this becomes a reality.”
Cara Lab
Thannhauser leads the Cara Lab -Cardiology Lab with Abbott, Radboudumc and Amsterdam UMC. Along with Niels Van Royen (Radboudumc) and Ivana Išgum (Amsterdam UMC), his team received a grant from the Netherlands Research Council (NWO). This study is one of the results.
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
