人工神经网络在体外受精胚胎评估中的应用  

Application of artificial neural network in evaluation of IVF embryo

作　　者：周德富[1] 邹廉[2] 陈瑛 Zhou Defu;Zou Lian;Chen Ying(Art College,Suzhou Vocational University,Jiangsu,Suzhou 215104,China;Reproduction Center,Wuxi Maternity and Child Health Care Hospital,Jiangsu,Wuxi 214002,China;Institute of Genetic Medicine,Wuxi Maternity and Child Health Care Hospital,Jiangsu,Wuxi 214002,China)

机构地区：[1]苏州市职业大学艺术学院,苏州215104 [2]无锡市妇幼保健院生殖中心,无锡214002 [3]无锡市妇幼保健院优生优育遗传医学研究所,无锡214002

出　　处：《中华检验医学杂志》2022年第3期310-314,共5页Chinese Journal of Laboratory Medicine

基　　金：江苏高校“青蓝工程”优秀教学团队(苏教师2018-12号);“太湖人才计划”顶尖医学专家团队项目(锡人才2021-9号)。

摘　　要：人工神经网络(ANN)是一种驱动人工智能(AI)的网络框架,其中采用经典卷积神经网络(CNN)进行胚胎质量评估可进行固定时间节点胚胎细胞计数和图像识别;采用全连接的深度神经网络(DNN),胚胎图像识别准确度提升,适用于较高硬件配置以及需要整合临床信息进行综合预测;残差网络通过增加层数提高准确度并通过跳跃连接解决梯度消失问题,实现动态胚胎评估。贝叶斯网络(BN)机器学习擅长推理,在条件缺失情况下可通过推理弥补数据不足,可结合临床复杂信息进行综合预测评估;支持向量机(MLP)机器学习存在梯度消失与爆炸,容易丢失图像部分空间特征,适用于小样本评估。ANN在预测胚胎植入率、胚胎非整倍体方面具有一定优势,开发新的胚胎质量评估方法减少侵入性检测是人类辅助生殖技术(ART)重要研究方向。Artificial neural network(ANN)is a network framework that drives artificial intelligence(AI).Classical convolutional neural networks(CNN)are mainly used for cell count and image recognition at fixed time in embryo evaluation.Fully connected deep neural networks(DNN),with increased accuracy of image recognition,are suitable for the units equipped with high configuration hardware and need comprehensive prediction according to the integrated clinical information.Residual networks improve the accuracy by increasing layers and solving the gradient disappearance problem through jump connection to realize dynamic embryo assessment.Bayesian networks(BN)and multi-layer perceptron(MLP)are two machine learning methods.The former is especially used for comprehensive prediction combined with complex clinical information in case of lack of conditions.The latter has gradient disappearance and explosion problem,and is easy to lose some spatial features of images,so it is used for small sample volumes.ANN has advantages in the prediction of implantation rate and aneuploidy and reducing invasive detection in quality assessment of embryos,which is an important research direction of human-assisted reproductive technology(ART).

关 键 词：人类辅助生殖技术 人工智能 深度学习 人工神经网络 

分 类 号：R714.8[医药卫生—妇产科学]