MEDICAL IMAGE CLASSIFICATION METHOD, APPARATUS AND DEVICE, AND STORAGE MEDIUM

摘要

The present application relates to the field of image classification. Disclosed are a medical image classification method, apparatus and device, and a storage medium. The method comprises: establishing an image library of relationship pairs; determining a pixel value of each picture included in each category of images in the image library, and calculating a prototype image corresponding to each category of images; acquiring an image to be classified, and respectively calculating the Euclidean distances between the image to be classified and each prototype image; sorting the Euclidean distances; and according to an obtained sorting result, determining the image category of the image to be classified. In the present application, GAN technology and data enhancement technology are used, so that existing image relationship pairs can be used to classify an image where the category of which is unknown, thereby solving the technical problem of it not being possible for data enhancement to generate high-quality scarce image data. Each category of images in an image library are all stored in a blockchain.