AI method automatically detects plaque erosion in heart arteries based on OCT images

Researchers have developed a new artificial intelligence (AI) method that can automatically detect plaque erosion in heart arteries using optical coherence tomography (OCT) images. It’s important to watch for plaque in the arteries because when plaque breaks down, it can block blood flow to the heart, leading to a heart attack or other serious conditions.

If the cholesterol plaque lining the arteries begins 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 new method could help improve the clinical diagnosis of plaque erosion and be used to develop new treatments for patients with heart disease. »

Zhao Wang, Research Team Leader, University of Electronic Science and Technology of China

OCT is an optical imaging method with micron-scale resolution that, when integrated with a miniaturized catheter, can be used in blood vessels to provide 3D images of the coronary arteries that supply blood to the heart . Although clinicians are increasingly using intravascular OCT to investigate plaque erosion, the large amount of data produced and the complexity of visual interpretation of images has led to significant interobserver variability.

To solve this problem, Wang worked with a group of engineers from his institution and doctors led by Bo Yu from the 2nd Affiliated Hospital of Harbin Medical University to develop an objective and automatic method that uses AI to detect the plaque erosion based on OCT images. They describe the new technique in the journal Optica Publishing Group Express Biomedical Optics and show that it is sufficiently accurate to potentially be used as a basis for clinical diagnosis.

“Our new AI-based method can automatically detect the presence of plaque erosion using the original OCT images without any additional input,” Wang said. “The ability to objectively and automatically detect plaque erosion will reduce the laborious manual assessment associated with diagnosis.”

Application of AI

The new method consists of two main steps. First, an AI model known as a neural network uses the original image and two shape information to predict regions of possible plaque erosion. The initial prediction is then refined with a post-processing algorithm based on clinically interpretable features that mimic the knowledge professional physicians use to make a diagnosis.

“We had to develop a new AI model that incorporates explicit shape information, the key feature used to identify plaque erosion in OCT images,” Wang said. “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.”

When OCT is used for intravascular imaging, the imaging probe is automatically pulled back inside a catheter, producing hundreds of images for each pullout. The researchers tested their method using 16 removals of 5,553 clinical OCT images with plaque erosion and 10 removals of 3,224 images without plaque erosion. The automated method correctly predicted 80% of plaque erosion cases with a positive predictive value of 73%. They also found that the diagnoses based on the automated method matched well with those of three experienced doctors.

“Although further safety validations and regulatory approvals are needed for stand-alone clinical use in patients, the technique could be used to aid in the diagnosis of plaque erosion,” Wang said. “This would involve doctors doing a final check on the results of the algorithm, then determining the cause of the acute coronary syndrome and the best treatment strategies.”

Investigate new treatments

The method could also be useful for analyzing the massive amounts of existing OCT data by eliminating the cumbersome and time-consuming process of manual image analysis. This could help scientists improve the identification and treatment of plaque erosion. For example, a stent is often used to recover reduced blood flow in patients with acute coronary syndrome, but recent studies suggest certain drugs may offer a less invasive alternative.

“Intravascular imaging, accompanied by AI technologies, can be an extremely valuable tool for coronary artery disease diagnosis and treatment planning,” Wang said. “In the future, this novel approach may help physicians develop individualized treatment strategies for optimal management of patients with acute coronary syndrome.”

The researchers are now working to improve their new technique by better incorporating 3D information and incorporating more unlabeled data to improve the performance of the AI ​​model. In the future, they also plan to use a larger dataset including a global population for training and evaluation of the algorithm. They also want to explore how it could be used in various clinical situations to further demonstrate its potential usefulness and value.