Integrated PET/MRI Radiomic Model Achieves Precise Prediction of High-risk Carotid Plaques

Mar 27, 2026

A substantial proportion of ischemic strokes are linked to atherosclerotic plaques in the carotid arteries, which supply blood to the brain. However, not all plaques carry the same danger. Some remain relatively stable, but the others become vulnerable to rupture and can trigger stroke. Early identification of high-risk carotid plaques is an important clinical goal.

Magnetic resonance vessel wall imaging (MR-VWI) can reveal plaque structure in great detail, helping clinicians evaluate features associated with vulnerability. However, structural imaging alone cannot fully capture the underlying inflammatory activity that often drives plaque instability and stroke risk.

In a study published in Journal of the American Heart Association, Prof. ZHANG Na's team from the Shenzhen Institutes of Advanced Technology of the Chinese Academy of Sciences, and the collaborators from Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, by combining [18F]fluorodeoxyglucose positron emission tomography (PET) and MR-VWI, built an integrated PET/MRI model to identify high-risk carotid plaques.

The integrated PET/MRI model integrates the assessments of plaque structure and metabolic activity in one framework. By employing detailed radiomic features from PET and MR-VWI, plaque characteristics such as surface ulceration and contrast enhancement, and a PET measure of metabolic activity, it can assess how fragile a plaque looks and how active its inflammation is, providing a more complete picture of stroke risk.

Researchers tested the model in a separate prospective patient group. The model showed strong and stable performance and worked better than models based on only one imaging method.

This integrated PET/MRI model offers a fast and fully non-invasive way to assess stroke risk, which helps doctors identify patients with high-risk carotid plaques more accurately and earlier.

Flow chart of the study.(Image by SIAT)



File Download:

    Media Contact
    YU Rong
    Email:
    rong.yu@siat.ac.cn
    Related Articles