Assessing plaque burden in a peripheral blood vessel of a subject

Abstract

A system and method are provided for assessing plaque burden in a peripheral blood vessel of a subject. The method includes acquiring a three-dimensional, 3D, ultrasound image of the vessel using non-invasive ultrasound imaging (S511); performing 3D vessel segmentation of the vessel in the ultrasound image using a vessel segmentation machine learning model to identify borders of the vessel (S512); performing 3D plaque segmentation of the plaque in the ultrasound image using a plaque segmentation machine learning model to identify plaque within the borders of the vessel (S513); detecting a centerline of the vessel based on the 3D vessel segmentation of the vessel (S516); determining volume of the plaque within the borders of the vessel based on the 3D vessel segmentation (S517); and determining local thickness of the plaque in at least one cross-section of the vessel relative to the detected centerline based on the 3D vessel segmentation and the 3D plaque segmentation (S518). Both 3D vessel segmentation and 3D plaque segmentation are performed for the entire image volume, as opposed to having to perform 2D vessel segmentation and 2D plaque segmentation of individual 2D cross-sections (frames) of the vessel.

Images