深度学习在骨伤影像学检查中的应用  

Application of deep learning in orthopedic trauma imaging

作　　者：谢秋晨 梅楠 陈坚[2] 尹波 XIE Qiuchen;MEI Nan;CHEN Jian;YIN Bo(Department of Radiology;Department of Gastroenterology,Huashan Hospital,Fudan University,Shanghai 200040,China)

机构地区：[1]复旦大学附属华山医院放射科,上海200040 [2]复旦大学附属华山医院消化内科,上海200040

出　　处：《上海医药》2020年第23期10-13,38,共5页Shanghai Medical & Pharmaceutical Journal

摘　　要：在过去的几年里,深度学习的发展势头很强劲。在骨科和创伤学领域,已有一些研究使用深度学习来辅助检测X线片中的骨折。相比之下,在通过CT检测骨折和进行骨折分类方面,使用深度学习的研究还较少。本文概要介绍深度学习用于X线片和CT图像上骨折检测的方法、深度学习对骨伤影像学诊断的赋能潜力以及深度学习在骨折检测中的偏差和未来发展方向。In the past few years,the development momentum of deep learning has been very strong.In the field of orthopedics and traumatology,there have been some studies using deep learning to detect fractures in X-rays imaging.In contrast,there are relatively few deep learning studies to detect and classify fractures by CT.In this narrative review,we give a brief overview of deep learning techniques,describe the methods that deep learning has been applied to fracture detection in X-rays and CT imaging so far,discuss how deep learning empowers this field and comment on the bias and future development direction of this technology.

关 键 词：深度学习 卷积神经网络 骨折检测 

分 类 号：TP391.5[自动化与计算机技术—计算机应用技术] R683[自动化与计算机技术—计算机科学与技术]