[1]侯鹏飞,沈秀明,袁明远,等.基于深度学习方法的胸片异物检测[J].中国医学物理学杂志,2021,38(12):1518-1523.[doi:DOI:10.3969/j.issn.1005-202X.2021.12.011]
 HOU Pengfei,SHEN Xiuming,et al.Detecting foreign objects in chest radiographs based on a deep learning method[J].Chinese Journal of Medical Physics,2021,38(12):1518-1523.[doi:DOI:10.3969/j.issn.1005-202X.2021.12.011]
点击复制

基于深度学习方法的胸片异物检测()
分享到:

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
38卷
期数:
2021年第12期
页码:
1518-1523
栏目:
医学影像物理
出版日期:
2021-12-24

文章信息/Info

Title:
Detecting foreign objects in chest radiographs based on a deep learning method
文章编号:
1005-202X(2021)12-1518-06
作者:
侯鹏飞14沈秀明2袁明远3孙九爱4
1.上海理工大学健康科学与工程学院, 上海 200093; 2.上海市松江区卫生人才培训中心, 上海 201600; 3.上海健康医学院附属周浦医院放射科, 上海 201318; 4.上海健康医学院医学影像学院, 上海 201318
Author(s):
HOU Pengfei1 4 SHEN Xiuming2 YUAN Mingyuan3 SUN Jiuai4
1. School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China 2. Healthcare Professional Training Center of Songjiang District, Shanghai 201600, China 3. Department of Radiology, Zhoupu Hospital Affiliated to Shanghai University of Medicine and Health Sciences, Shanghai 201318, China 4. School of Medical Imaging, Shanghai University of Medicine and Health Sciences, Shanghai 201318, China
关键词:
胸片异物检测深度学习YOLO网络
Keywords:
Keywords: chest radiograph foreign object detection deep learning YOLO network
分类号:
R318;TP18
DOI:
DOI:10.3969/j.issn.1005-202X.2021.12.011
文献标志码:
A
摘要:
为了提高胸片异物自动检测的能力,采用深度学习网络高效提取各种尺度和形状的异物影像特征,实现胸片中多种异物的自动稳定检测。在网络构建过程中,根据异物的影像特征,改进YOLO v4目标检测网络,通过在特征提取网络CSPDarkNet53中加入SE-block(Squeeze and Excitation),使模型能够区别利用各个通道的信息。实验结果表明,改进的深度学习网络在异物检测中能够实现92%的精确率和83%的召回率。因此,新的深度学习方法可用于胸片异物检测等应用场景,客观评判摄影质量,为放射影像的质量控制打下基础。
Abstract:
Abstract: In order to improve the ability of automatically detecting foreign objects in chest radiographs, a deep learning network is used to capture the imaging features of foreign objects with various scales and shapes efficiently, thereby realizing the automatic and stable detection of various foreign objects in chest radiographs. During network construction, according to the imaging features of foreign objects, YOLO v4 network is improved by embedding SE-block (Squeeze and Excitation) into the feature extraction network CSPDarkNet53 to make the model capable of distinguishing and utilizing the information of each channel. Experimental results demonstrate that the proposed deep learning network achieves 92% accuracy and 83% recall rate for foreign object detection. Therefore, the proposed deep learning approach can be used to identify foreign objects in chest radiographs and assess image quality automatically and objectively, which lays a foundation for the quality control of radiographic images.

备注/Memo

备注/Memo:
【收稿日期】2021-07-14 【基金项目】上海健康医学院协同创新重点专项(SPCI-18-17-001) 【作者简介】侯鹏飞,硕士,研究方向:医学图像处理技术,E-mail: 892794685@qq.com 【通信作者】孙九爱,博士,副教授,研究方向:医学成像与图像处理技术,E-mail: sunja@sumhs.edu.cn
更新日期/Last Update: 2021-12-24