摘要
基于算法学习数据内部规律,进而对同类数据进行预测和判断的过程为机器学习。在体外受精-胚胎移植技术领域,基于机器学习算法建立的模型不仅可预测周期助孕结局,也可帮助胚胎学家挑选优质胚胎。本文共筛选出基于机器学习算法的周期结局预测模型和胚胎质量评估模型30个,其中基于传统机器学习算法模型28个,基于深度学习模型2个。采用受试者工作特征曲线的曲线下面积(area under curve,AUC)评价模型效果,基于传统机器学习算法的模型效果多不理想(0.60<AUC<0.86),深度学习算法准确率则较高(AUC>0.90)。完善的预测和评估模型有望提高助孕周期效率、标准化胚胎选择流程。
Based on algorithm,machine learning could dig information from data and learn the rules between data,following by predicting and analyzing new data.Machine learning can be used for the establishment of pregnancy outcome prediction model,as well as for the selection of embryos with the highest implantation potential.This review identified 30 models,among which 28 were based on traditional machine learning and 2 were based on deep learning.Area under the receiver operating characteristic curve(AUC)was adopted for the estimation of model performance.On the whole,models based on traditional algorithm were of low to medium performance(0.60<AUC<0.86),whereas deep learning models were of good performance(AUC>0.90).Prediction and estimation models may improve treatment efficiency and standardize embryo selection process.
作者
于医萍
高一博
方兰兰
孙莹璞
Yu Yiping;Gao Yibo;Fang Lanlan;Sun Yingpu(The Reproductive Medicine Center of the First Affiliated Hospital of Zhengzhou University,Henan Province Key Laboratory of Reproduction and Genetics,Zhengzhou 450052,China)
出处
《中华生殖与避孕杂志》
CSCD
北大核心
2021年第10期883-892,共10页
Chinese Journal of Reproduction and Contraception
关键词
机器学习
统计模型
受精
体外
周期结局
胚胎质量
Machine learning
Statistical models
Fertilization in vitro
Pregnancy outcomes
Embryo quality
