PLAQUE PROCESSING METHOD AND DEVICE CAPABLE OF PERFORMING MAGNETIC RESONANCE VESSEL WALL IMAGING, AND COMPUTING DEVICE

摘要

A plaque processing method and device capable of performing magnetic resonance vessel wall imaging, and a computing device, being used to efficiently and accurately identify and classify plaques in magnetic resonance vessel wall imaging, belonging to the field of image processing. Said method comprises: training a generative adversarial network and a capsule neural network, so as to obtain a trained generator network and a trained capsule neural network (S101); and cascading the trained generator network and capsule neural network into a system, so as to identify and classify plaques in the magnetic resonance vessel wall imaging (S102). The present invention enables the vessel plaque feature information, represented by a vector, in the capsule neural network to be richer; and greatly improves the accuracy of identifying and classifying when the trained generator network and capsule neural network are cascaded into a system for identifying and classifying plaques in the magnetic resonance vessel wall imaging.