Technology Description
Researchers at Washington University in St. Louis have developed a two-stage neural network model, with CNN and BNN architecture, to segment carotid atherosclerotic plaque components based on multi-weighted MR images and measure the uncertainty of the segmentation output. This model identifies the lipid-rich necrotic core of the carotid atheroma for use in determining the plaque’s vulnerability to rupture and cause ischemic stroke.
Stage of Research
Researchers have trained the networks using high-resolution MRI ex vivo data, as well as pathology samples of the same plaque obtained from patients post-surgery.
Publications
– Li R, Zheng J, Zayed MA… Jha AK. (2023). Carotid atherosclerotic plaque segmentation in multi-weighted MRI using a two-stage neural network: advantages of training with high-resolution imaging and histology. Frontiers in Cardiovascular Medicine, 10:1127653.
Applications
– Diagnostic imaging for potential stroke risk
Key Advantages
– Reliable and automated segmentation method
Patents: Pending
