Chromosome segmentation recognition method and device based on graph convolutional neural network, and electronic equipment

• Main Authors: CAI YUFENG, LIU XIANGYONG, MU YANG, PENG WEIXIONG, LU QINYANG
• Format: Patent
• Language: Chinese; English
• Published: 28.01.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides a chromosome segmentation recognition method and device based on a graph convolutional neural network, and electronic equipment. The method comprises the following steps that: each chromosome sub-graph of a chromosome metaphase graph and a chromosome segmentation identification model based on a Unet neural network are provided; each chromosome sub-graph is input into an encoder and an intersection-to-union ratio calculation unit, the encoder performs encoding processing according to each chromosome sub-graph to obtain a feature vector, the intersection-to-union ratio calculation unit sets each chromosome sub-graph as a rectangular frame in a chromosome metaphase graph, and the intersection-to-union ratio of each rectangular frame is obtained through calculation; the adjacent matrix input unit generates an adjacent matrix according to an intersection-to-parallel ratio; a graph convolution unit performs graph convolution on the feature vector according to the adjacent matrix; and the deco