深度学习用于产前超声显示切面分析研究进展  

Research progresses of deep learning applicated in prenatal ultrasonic section analysis

作　　者：高云玥 福林[1] 王宗敏 GAO Yunyue;FU Lin;WANG Zongmin(Department of Ultrasound Diagnosis,Affiliated Hospital of Inner Mongolia Medical University,Hohhot 010050,China)

机构地区：[1]内蒙古医科大学附属医院超声诊断科,内蒙古呼和浩特010050

出　　处：《中国医学影像技术》2023年第12期1902-1905,共4页Chinese Journal of Medical Imaging Technology

基　　金：内蒙古医科大学科技百万工程联合项目[YKD2018KJBW(LH)036]。

摘　　要：超声已成为产前检查的首选影像学方法,但其操作者依赖性和成像高变异性影响结果的准确性及检查效率。深度学习(DL)用于图像分割与识别时,可于较低级别特征中派生出较高级别特征,由此准确描述原始数据中的固有规律或信息特征。本文就DL用于产前超声不同切面分析的研究进展进行综述。Ultrasound has become the optimal imaging method for prenatal screening,whereas the operator dependence and high imaging variability may affect the accuracy of results and examination efficiency.Applied in segmentation and recognition tasks in images,deep learning(DL)is able to derive higher-level features from lower-level features,accurately characterize the inherent laws or information features of raw data.The research progresses of DL applicated in prenatal ultrasonic section analysis were reviewed in this article.

关 键 词：深度学习 超声检查 产前 

分 类 号：R714[医药卫生—妇产科学] R445.1[医药卫生—临床医学]