基于YOLO的染色体着丝粒识别方法研究  

Study on Chromosome Centromere Recognition Method Based on YOLO

作　　者：董娟聪[1] 蒋文轩 原雅艺 党旭红[1] Dong Juancong;Jiang Wenxuan;Yuan Yayi;Dang Xuhong(China Institute for Radiation Protection,Taiyuan,Shanxi 030006;Guangxi University of Science and Technology,Liuzhou,Guangxi 545006)

机构地区：[1]中国辐射防护研究院,太原030006 [2]广西科技大学,广西柳州545006

出　　处：《辐射防护通讯》2022年第4期65-68,共4页Radiation Protection Bulletin

摘　　要：快速、准确识别染色体上的着丝粒,是实现染色体畸变自动化检测分析技术的关键环节之一。本研究利用计算机深度学习对荧光探针标记的染色体畸变图片着丝粒进行分析,初步实现了自动识别图像中的着丝粒,可为自动化检测染色体畸变系统的研发奠定一定的研究基础。Rapid and accurate identification of centromere on chromosomes is one of the key steps to realize automatic analysis of chromosome aberrations.We employ computer deep learning to analyze chromosomal aberration images labeled by fluorescent probes.It realizes automatic recognition of centromere in the image,and is expected to provide technical support for the development of automatic determination system of chromosomal aberration.

关 键 词：YOLO 染色体着丝粒 自动识别 

分 类 号：R148[医药卫生—公共卫生与预防医学]