Method and system for intelligent prediction of embryo pregnancy status

Abstract

The invention discloses a method and system for intelligently predicting the pregnancy state of an embryo. The method comprises: collecting embryo images within D1 to D6; inputting the embryo images into a fragment ratio prediction network model, and a blastocoel and inner cell mass grade prediction network Model, trophoblast grade prediction network model calculates and outputs the predicted fragment ratio of the embryo image, the predicted blastocoel ratio and the predicted inner cell mass grade, and the predicted trophoblast grade; then input to the embryo pregnancy rate state prediction machine learning model to calculate and output the embryo Pregnancy rate prediction results. The intelligent prediction method and system proposed by the present invention monitor the whole process of embryo development, and use the comprehensive scoring function to calculate the pregnancy rate of embryos, without manual intervention in the prediction process, which can help doctors quickly and accurately make judgments on embryo scoring.

Description

technical field [0001] The present invention relates to the technical fields of embryo morphology and artificial intelligence, in particular to an intelligent prediction method and system of embryo pregnancy state. Background technique [0002] The quality of embryo development directly affects the result of the pregnancy rate. Embryologists rely on two main ways of embryo morphology and genetics to judge the quality of embryos. Among them, the use of genetic methods to judge the quality of embryos requires extremely high experimental conditions. , and the use of embryo morphological information to complete the evaluation of embryos is a simple, fast and effective method. At present, most embryologists, based on their own long-term embryo observation experience, complete the judgment of the quality of the embryo by obtaining some important morphological characteristics of the embryo during the development process. in accordance with. Doctors usually choose the period of D1...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the problems of low accuracy and poor real-time performance of traditional image analysis methods in the prior art in the acquisition of fragmented regions, calculation of blastocoel ratio, evaluation of inner cell mass and trophoblast grade, A method and system for intelligent detection of embryo pregnancy status is proposed, which can automatically complete the extraction and prediction of high-level features of fragments, blastocoel, inner cell mass and trophoblast, and realize the comprehensive prediction of embryo pregnancy status accurately and efficiently.

Images