In what could lead to the development of new treatments for heart disease, researchers have developed a new artificial intelligence (AI) technique that can detect plaque erosion in heart arteries. The technique uses optical coherence tomography (OCT) images to track arterial plaque. The finding is crucial because the disintegration of the plaque can serve as a prelude to a heart attack or other serious heart disease. The OCT, an optical imaging technique, can be used in blood vessels to create 3D images of the coronary arteries that carry blood to the heart muscles.
The OCT technique has been in use to detect plaque erosion. But the amount of data generated and the difficulty of interpreting the images causes significant inter-observer variability. To solve this, researchers introduced AI into the OCT technique to identify plaque erosion.
“When cholesterol plaques lining the arteries begin to erode, it can lead to a sudden reduction in blood flow to the heart, known as acute coronary syndrome, requiring urgent treatment. Our new method could help improve the clinical diagnosis of plaque erosion and can be used to develop new treatments for patients with heart disease,” said Zhao Wang of the University of Electronic Science and Technology of China, who is also the lead author of the study published in Biomedical Optics Express.
The new AI method consists of two primary steps, where first an AI model known as a neural network analyzes the original image and two pieces of shape information to predict the regions of plaque erosion.
This prediction is then refined by a post-processing algorithm that mimics the knowledge of professional doctors to make the diagnosis. “We needed to develop a new AI model that contains explicit shape information, the key feature used to identify plaque erosion in OCT images. The underlying intravascular OCT imaging technology is also crucial as it is currently the highest-resolution imaging modality.” that can be used to diagnose plaque erosion in living patients,” Wang said.
The researchers tested the new technique and found that the automated method could predict 80 percent of plaque erosion cases with a positive predictive value of 73 percent. In addition, the diagnosis made using AI technique matched well with that of experienced physicians.