正在加载图片...
第12卷第6期 智能系统学报 Vol.12 No.6 2017年12月 CAAI Transactions on Intelligent Systems Dec.2017 D0:10.11992/tis.201706035 网络出版地址:http:/kns.cnki.net/kcms/detail/23.1538.TP.20171109.1250.014.html 基于医学征象和卷积神经网络的肺结节 CT图像哈希检索 杨晓兰',强彦,赵涓涓,杜晓平2,赵文婷 (1.太原理工大学计算机科学与技术学院,山西太原030024:2.山西省煤炭中心医院PET/CT中心,山西太原030012) 摘要:针对肺结节图像检索中存在的两个问题:手工设计的特征对肺结节的表达能力不强,生成的哈希码检索效果 不佳。文中提出一种基于医学征象和卷积神经网络的肺结节CT图像哈希检索方法。首先,依据肺结节的9种征象 取值,构造训练集准确的哈希码:其次,利用卷积神经网络和主成分分析法提取肺结节的重要语义特征,并结合训练 集准确的哈希码反向求解哈希函数:最后,提出一种基于自适应比特位的检索方法,实现待查询肺结节图像的快速检 出。通过对数据集进行实验和分析,证实了本文方法在肺结节图像检索过程中取得了较高的准确率和检索精度。 关键词:肺结节;医学征象;卷积神经网络;主成分分析;语义特征;哈希函数;自适应:图像检索 中图分类号:TP391文献标志码:A 文章编号:1673-4785(2017)06-0857-08 中文引用格式:杨晓兰,强彦,赵涓涓,等.基于医学征象和卷积神经网络的肺结节CT图像哈希检索J几.智能系统学报,2017,12(6: 857-864. 英文引用格式:YANG Xiaolan,.QIANG Yan,ZHAO Juanjuan,,etal.Hashing retrieval for CT images of pulmonary nodules based on medical signs and convolutional neural networks[J.CAAI transactions on intelligent systems,2017,12(6):857-864. Hashing retrieval for CT images of pulmonary nodules based on medical signs and convolutional neural networks YANG Xiaolan',QIANG Yan',ZHAO Juanjuan',DU Xiaoping,ZHAO Wenting (1.College of Computer Science and Technology,Taiyuan University of Technology,Taiyuan 030024,China;2.PET/CT Center of Shanxi Coal Central Hospital,Taiyuan 030012,China) Abstract:Existing pulmonary nodule retrieval methods have two problems;it is difficult to express the characteristics of pulmonary nodules using hand-crafted features and the generated hashing codes have poor retrieval performance.To address these issues,this paper proposes a retrieval method for pulmonary nodules in CT images based on medical signs and convolutional neural networks.We first constructed accurate hashing codes using an accurate training set based on the values of the nine signs of pulmonary nodules.We then extracted the important semantic features of pulmonary nod- ules using convolutional neural networks and principal components analysis.In addition,we inversely solved the hash- ing functions by combining the hashing codes with the accurate training set.Finally,we developed a retrieval method, based on adaptive bits,to achieve fast searching for pulmonary nodule images.Extensive experiments and evaluations on data sets show that the method has high accuracy and retrieval precision in the process of pulmonary nodule image retrieval. Keywords:pulmonary nodules:medical signs:convolutional neural networks:principal components analysis:semantic features;Hashing Function;adaptive;image retrieval 收稿日期:2017-06-13.网络出版日期:2017-11-09 肺癌是目前世界上发病率最高的恶性肿瘤之 基金项目:国家自然科学基金项目(61373100):虚拟现实技术与系 统国家重点实验室开放基金项目(BUAA-VR-I7KF-14, 一,也是男女癌症死亡的主要原因。肺癌的早期 BUAA-VR-17KF.15):山西省回国留学人员科研项目 (2016-038). 检测和诊断在提高治愈率方面起着重要的作用。薄 通信作者:杨晓兰.E-mail:1141183481@qq.com 扫CT作为早期肺癌筛查的重要手段,可以大幅度DOI: 10.11992/tis.201706035 网络出版地址: http://kns.cnki.net/kcms/detail/23.1538.TP.20171109.1250.014.html 基于医学征象和卷积神经网络的肺结节 CT 图像哈希检索 杨晓兰1 ,强彦1 ,赵涓涓1 ,杜晓平2 ,赵文婷1 (1. 太原理工大学 计算机科学与技术学院,山西 太原 030024; 2. 山西省煤炭中心医院 PET/CT 中心,山西 太原 030012) 摘 要:针对肺结节图像检索中存在的两个问题:手工设计的特征对肺结节的表达能力不强,生成的哈希码检索效果 不佳。文中提出一种基于医学征象和卷积神经网络的肺结节 CT 图像哈希检索方法。首先,依据肺结节的 9 种征象 取值,构造训练集准确的哈希码;其次,利用卷积神经网络和主成分分析法提取肺结节的重要语义特征,并结合训练 集准确的哈希码反向求解哈希函数;最后,提出一种基于自适应比特位的检索方法,实现待查询肺结节图像的快速检 出。通过对数据集进行实验和分析,证实了本文方法在肺结节图像检索过程中取得了较高的准确率和检索精度。 关键词:肺结节;医学征象;卷积神经网络;主成分分析;语义特征;哈希函数;自适应;图像检索 中图分类号:TP391 文献标志码:A 文章编号:1673−4785(2017)06−0857−08 中文引用格式:杨晓兰, 强彦, 赵涓涓, 等. 基于医学征象和卷积神经网络的肺结节 CT 图像哈希检索[J]. 智能系统学报, 2017, 12(6): 857–864. 英文引用格式:YANG Xiaolan, QIANG Yan, ZHAO Juanjuan, et al. Hashing retrieval for CT images of pulmonary nodules based on medical signs and convolutional neural networks[J]. CAAI transactions on intelligent systems, 2017, 12(6): 857–864. Hashing retrieval for CT images of pulmonary nodules based on medical signs and convolutional neural networks YANG Xiaolan1 ,QIANG Yan1 ,ZHAO Juanjuan1 ,DU Xiaoping2 ,ZHAO Wenting1 (1. College of Computer Science and Technology, Taiyuan University of Technology, Taiyuan 030024, China; 2. PET/CT Center of Shanxi Coal Central Hospital, Taiyuan 030012, China) Abstract: Existing pulmonary nodule retrieval methods have two problems; it is difficult to express the characteristics of pulmonary nodules using hand-crafted features and the generated hashing codes have poor retrieval performance. To address these issues, this paper proposes a retrieval method for pulmonary nodules in CT images based on medical signs and convolutional neural networks. We first constructed accurate hashing codes using an accurate training set based on the values of the nine signs of pulmonary nodules. We then extracted the important semantic features of pulmonary nod￾ules using convolutional neural networks and principal components analysis. In addition, we inversely solved the hash￾ing functions by combining the hashing codes with the accurate training set. Finally, we developed a retrieval method, based on adaptive bits, to achieve fast searching for pulmonary nodule images. Extensive experiments and evaluations on data sets show that the method has high accuracy and retrieval precision in the process of pulmonary nodule image retrieval. Keywords: pulmonary nodules; medical signs; convolutional neural networks; principal components analysis; semantic features; Hashing Function; adaptive; image retrieval 肺癌是目前世界上发病率最高的恶性肿瘤之 一,也是男女癌症死亡的主要原因[1]。肺癌的早期 检测和诊断在提高治愈率方面起着重要的作用。薄 扫 CT 作为早期肺癌筛查的重要手段,可以大幅度 收稿日期:2017−06−13. 网络出版日期:2017−11−09. 基金项目:国家自然科学基金项目(61373100);虚拟现实技术与系 统国家重点实验室开放基金项目(BUAA-VR-17KF-14, BUAA-VR-17KF-15);山西省回国留学人员科研项目 (2016-038). 通信作者:杨晓兰. E-mail:1141183481@qq.com. 第 12 卷第 6 期 智 能 系 统 学 报 Vol.12 No.6 2017 年 12 月 CAAI Transactions on Intelligent Systems Dec. 2017
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有