A Method for Embryo Tissue Segmentation Based on Generative Adversarial Networks

Abstract

The invention relates to an embryo tissue segmentation method based on a generative confrontation network, which belongs to the technical field of medical image processing. Including: step 1 to perform tissue segmentation mask mapping on embryonic tissue slice images through the U‑net network; step 2 to make a segmentation network training set; step 3 to configure the parameters required for network training to obtain the set network; step 4 to use the completed The organizational quality recognition network after the training and setting of the segmentation network training set; step 5 fixes the parameters of the organizational quality recognition network, and uses the completed segmentation network training set combined with the U-net network after the organizational quality recognition network training setting; step 6 will be unmarked The embryonic tissue slice image of the segmentation result is used as input, and the corresponding mask image is generated. The network based on the segmentation method uses a classification model to supplement the loss during training segmentation, fully utilizes the information of cell growth status, and improves the accuracy of the segmentation network in the field of embryonic tissue segmentation.

Description

Technical field [0001] The present invention relates to an embryonic tissue segmentation method based on a generating counterfeit network, which belongs to the technical field of medical image processing. Background technique [0002] With the rapid development of digital imaging technology, medical imaging is gradually widely used in clinical detection and treatment. The doctor can use medical imaging to intervene in judging pathological reasons, the treatment process, and the effect is accurately diagnosed and caused by the effect after treatment, so intelligent diagnosis reduces the diagnosis time while improving the accuracy of the diagnosis. It is probably divided into three processes for the observation and classification of medical images. It starts to observe the image from experienced physicians, and the subjective factors are strong and cost-effective; the identification of images after the addition of computer technology can do semi-automatic work, that is, participati...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a method for embryonic tissue segmentation based on generative adversarial networks for the lack of data samples in the training process of the existing medical assistance system, which results in the use of traditional machine learning methods for training. Neural Network Method for Medical Assisted Diagnosis and Tissue Segmentation Recognition

Images