DEEP LEARNING ARCHITECTURE SYSTEM FOR AUTOMATIC INTERPRETATION OF FUNDUS IMAGE AND METHOD FOR AUTOMATIC INTERPRETATION OF FUNDUS IMAGE THEREOF

摘要

The present invention relates to an algorithm for automatic reading of the fundus image, a dip for the automatic reading of the fundus image that can minimize the amount of data required for learning by training and reading artificial intelligence in a manner similar to that of an ophthalmologist acquiring medical knowledge. It's about running architecture. The deep learning architecture system for automatic reading of the fundus image according to the present invention is composed of a plurality of convolutional layers performing feature extraction of the fundus image and one fulling layer performing subsampling to reduce computation amount A trunk module 100 in which a common part is combined in a plurality of convolutional neural network (CNN) architectures in which at least one extraction layer set is arranged in series; A branch module 200 configured to generate a plurality of architectures in the trunk module 100 and receiving the output of the trunk module 100 to identify lesions of the fundus image and diagnose a corresponding disease name; A section 110 which is an architecture in which any one of the branch modules 200 and the trunk module 100 are connected; A root layer 120 for transmitting the output of a specific layer among the trunk modules 100 to the branch module 200 to connect the trunk module 100 and the branch module 200; And a final diagnosis unit 300 for integrating data diagnosed from the branch module 200 and determining and outputting a final disease name.