A novel method makes it simpler to spot plaque degradation by using intravascular optical coherence tomography pictures.
Researchers have developed a brand-new artificial intelligence (AI) method that automatically detects plaque erosion in the heart’s arteries using optical coherence tomography (OCT) pictures. It is critical to monitor arterial plaque because, if it breaks down, it may impede blood flow to the heart, resulting in a heart attack or other serious issues.
According to Zhao Wang, the head of the research team from the University of Electronic Science and Technology of China, “if cholesterol plaque lining arteries starts to erode it can lead to a sudden reduction in blood flow to the heart known as acute coronary syndrome, which requires urgent treatment.”
“Our innovative approach could aid in improving the clinical diagnosis of plaque erosion and be used to create fresh therapies for heart disease patients,” according to the study.
OCT is a micron-scale resolution optical imaging technique that can be used inside blood vessels to create 3D images of the coronary arteries, which carry blood to the heart. Despite the fact that doctors are using intravascular OCT more often to look for plaque erosion, there is a significant amount of interobserver variability due to the volume of data generated and the challenge of visually interpreting the pictures.
In order to address this issue, Wang collaborated with a team of engineers from his institution and medical professionals from The 2nd Affiliated Hospital of Harbin Medical University under the direction of Bo Yu to create an automated, objective method that uses AI to identify plaque erosion based on OCT images. They explain the new method in the Optica Publishing Group journal Biomedical Optics Express and demonstrate that it is accurate enough to possibly serve as the foundation for clinical diagnosis.
There are two main steps in the new methodology. First, regions of potential plaque erosion are predicted using the original image and two pieces of shape information using an AI model known as a neural network. A post-processing algorithm built on clinically interpretable information that mimics the knowledge used by trained medical professionals to make a diagnosis then refines the initial prediction.