Article(id=1327013128828727450, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327013054811845282, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202440527, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1718553600000, receivedDateStr=2024-06-17, revisedDate=1740672000000, revisedDateStr=2025-02-28, acceptedDate=1743609600000, acceptedDateStr=2025-04-03, onlineDate=1794314184624, onlineDateStr=2026-11-10, pubDate=1751299200000, pubDateStr=2025-07-01, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794314184624, onlineIssueDateStr=2026-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794314184624, creator=13701087609, updateTime=1794314184624, updator=13701087609, issue=Issue{id=1327013054811845282, tenantId=1146029695717560320, journalId=1324759914146529478, year='2025', volume='62', issue='7', pageStart='1595', pageEnd='1832', issueExtLink='null', onlineDate='null', pubDate='1751299200000', pubDateStr='2025-07-01', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794314166977, creator='13701087609', updateTime=1794376853183, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1327275980453950013, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327013054811845282, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1327275980453950014, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327013054811845282, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=1622, endPage=1638, ext={EN=ArticleExt(id=1327013129202020507, articleId=1327013128828727450, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=ECG-QGAN: An ECG Generative Information System Based on Quantum Generative Adversarial Networks, columnId=1327013055591985830, journalTitle=Journal of Computer Research and Development, columnName=Special Issue on Generative AI-driven Information Systems, runingTitle=null, highlight=null, articleAbstract=

According to statistics, the number of people suffering from cardiovascular diseases in China is about 330 million, and the number of deaths caused by cardiovascular diseases accounts for 40% of the total number of deaths each year. Under this circumstance, the development of heart disease assisted diagnosis systems is particularly important, but its development is limited by the lack of a large amount of electrocardiogram (ECG) clinical data that don’t contain patient privacy information and need to be annotated by medical experts. As an emerging discipline, quantum computing can explore larger and more complex state spaces by utilizing quantum superposition and entanglement properties, which is beneficial for generating high-quality and diverse electrocardiogram data similar to real clinical data. Therefore, we propose an electrocardiogram generative information system based on quantum generative adversarial networks, abbreviated as ECG-QGAN. The quantum generative adversarial network consists of a quantum bidirectional gated recurrent unit (QBiGRU) and a quantum convolutional neural network (QCNN). The system utilizes the entanglement property of quantum to improve the generative capability to produce ECG data that are consistent with the existing clinical data so that the heartbeat characteristics of cardiac patients can be preserved. The generator and discriminator of this system use QBiGRU and QCNN, respectively. The variational quantum circuit (VQC) designed based on matrix product state (MPS) and tree tensor network (TTN) is adopted, which enables the system to capture ECG data information more efficiently and generate qualified ECG data with fewer quantum resources. In addition, the system applies quantum Dropout technology to avoid overfitting issues during the training process. Finally, the experimental results show that the ECGs generated by ECG-QGAN have a higher average classification accuracy compared with other models for generating ECGs. It is also friendly to the current noise intermediate scale quantum (NISQ) computers in terms of the number of quantum bits and circuit depth.

, authors=Zhiguo Qu1, Weilong Chen1, Le Sun1, *, Wenjie Liu2, Yanchun Zhang3, 4, authorsList=Zhiguo Qu, Weilong Chen, Le Sun, Wenjie Liu, Yanchun Zhang, authorCompany=null, correspAuthors=Le Sun, authorNote=

born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

born in 2000. Master. His main research interest includes quantum neural networks

born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

, correspAuthorsNote=null, copyrightStatement=Copyright © 2025 Journal of Computer Research and Development. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1327013135933878476, articleId=1327013128828727450, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=ECG-QGAN:基于量子生成对抗网络的心电图生成式信息系统, columnId=1327013055789118120, journalTitle=计算机研究与发展, columnName=生成式人工智能驱动的信息系统专题, runingTitle=null, highlight=null, articleAbstract=

据统计,我国心血管疾病患病人数约达3.3亿,每年因为心血管疾病死亡的人数占总死亡人数的40%. 在这种背景下,心脏病辅助诊断系统的发展显得尤为重要,但其开发受限于缺乏不含患者隐私信息和由医疗专家标注的大量心电图(electrocardiogram, ECG)临床数据. 作为一门新兴学科,量子计算可通过利用量子叠加和纠缠特性,能够探索更大、更复杂的状态空间,进而有利于生成同临床数据一样的高质量和多样化的ECG数据. 为此,提出了一种基于量子生成对抗网络(QGAN)的ECG生成式信息系统,简称ECG-QGAN. 其中QGAN由量子双向门控循环单元(quantum bidirectional gated recurrent unit, QBiGRU)和量子卷积神经网络(quantum convolutional neural network, QCNN)组成. 该系统利用量子的纠缠特性提高生成能力,以生成与现有临床数据一致的ECG数据,从而可以保留心脏病患者的心跳特征. 该系统的生成器和判别器分别采用QBiGRU和 QCNN,并应用了基于矩阵乘积状态(matrix product state, MPS)和树形张量网络(tree tensor network, TTN)所设计的变分量子电路(variational quantum circuit , VQC),可以使该系统在较少的量子资源下更高效地捕捉ECG数据信息,生成合格的ECG数据. 此外,该系统应用了量子Dropout技术,以避免训练过程中出现过拟合问题. 最后,实验结果表明,与其他生成ECG数据的模型相比,ECG-QGAN生成的ECG数据具有更高的平均分类准确率. 同时它在量子位数量和电路深度方面对当前噪声较大的中尺度量子(noise intermediate scale quantum, NISQ)计算机是友好的.

, authors=瞿治国1, 陈韦龙1, 孙乐1, *, 刘文杰2, 张彦春3, 4, authorsList=瞿治国, 陈韦龙, 孙乐, 刘文杰, 张彦春, authorCompany=null, correspAuthors=孙乐, authorNote=

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

2000年生. 硕士. 主要研究方向为量子神经网络

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

, correspAuthorsNote=
孙乐()
, copyrightStatement=版权所有 © 2025 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=cedMkfjLcdldJMsuOmyuOA==, magXml=zZYUc5GgbxaLJwgQcqj7pg==, pdfUrl=null, pdf=Hem0/GCB6JIudedky+rQKw==, pdfFileSize=5060948, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=YNMZW+MD/UmC0+CSgW5KYw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=QLmZb8UBCsSMhpQpA88L0A==, mapNumber=null, fund=null)}, authors=[Author(id=1327013136634327259, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013136734990557, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136634327259, language=EN, stringName=Zhiguo Qu, firstName=Zhiguo, middleName=null, lastName=Qu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

"}, bioImg=null, bioContent=

born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013136839848158, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136634327259, language=CN, stringName=瞿治国, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"z3C66fWIRK5JM6qRro/fIg==","content":"

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

"}, bioImg=z3C66fWIRK5JM6qRro/fIg==, bioContent=

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013136940511456, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137028591842, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136940511456, language=EN, stringName=Weilong Chen, firstName=Weilong, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 2000. Master. His main research interest includes quantum neural networks

"}, bioImg=null, bioContent=

born in 2000. Master. His main research interest includes quantum neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137137643748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136940511456, language=CN, stringName=陈韦龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"tscjdL/Opx2ZOLF+bzNYGQ==","content":"

2000年生. 硕士. 主要研究方向为量子神经网络

"}, bioImg=tscjdL/Opx2ZOLF+bzNYGQ==, bioContent=

2000年生. 硕士. 主要研究方向为量子神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013137229918438, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=002359@nuist.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327013137317998824, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137229918438, language=EN, stringName=Le Sun, firstName=Le, middleName=null, lastName=Sun, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

"}, bioImg=null, bioContent=

born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137406079209, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137229918438, language=CN, stringName=孙乐, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"fQSakqWUzx13IgoxVJhigw==","content":"

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

"}, bioImg=fQSakqWUzx13IgoxVJhigw==, bioContent=

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013137502548203, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137582239981, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137502548203, language=EN, stringName=Wenjie Liu, firstName=Wenjie, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137670320366, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137502548203, language=CN, stringName=刘文杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2南京信息工程大学软件工程学院 南京 210044, bio={"img":"qvghsECXHdxeTrw9TzF3sg==","content":"

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

"}, bioImg=qvghsECXHdxeTrw9TzF3sg==, bioContent=

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136298782929, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=2, ext=[AuthorCompanyExt(id=1327013136307171538, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136332337363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学软件工程学院 南京 210044)])]), Author(id=1327013137787760880, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137917784307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137787760880, language=EN, stringName=Yanchun Zhang, firstName=Yanchun, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004
4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066, bio={"content":"

born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

"}, bioImg=null, bioContent=

born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137997476084, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137787760880, language=CN, stringName=张彦春, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3浙江师范大学计算机科学与技术学院 浙江金华 321004
4鹏城实验室新型网络研究部 广东深圳 518066, bio={"img":"IOt9r0FLojX/7aQzdnXNxg==","content":"

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

"}, bioImg=IOt9r0FLojX/7aQzdnXNxg==, bioContent=

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136437194964, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=3, ext=[AuthorCompanyExt(id=1327013136445583573, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004), AuthorCompanyExt(id=1327013136458166486, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3浙江师范大学计算机科学与技术学院 浙江金华 321004)]), AuthorCompany(id=1327013136537858263, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=4, ext=[AuthorCompanyExt(id=1327013136550441176, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066), AuthorCompanyExt(id=1327013136558829785, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4鹏城实验室新型网络研究部 广东深圳 518066)])])], keywords=[Keyword(id=1327013138215579893, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, orderNo=1, keyword=generative information system), Keyword(id=1327013138307854582, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, orderNo=2, keyword=electrocardiogram (ECG)), Keyword(id=1327013138404323575, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, orderNo=3, keyword=quantum generative adversarial network (QGAN)), Keyword(id=1327013138471432440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, orderNo=4, keyword=quantum bidirectional gated recurrent unit (QBiGRU)), Keyword(id=1327013138542735609, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, orderNo=5, keyword=quantum convolutional neural network (QCNN)), Keyword(id=1327013140212068602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, orderNo=1, keyword=生成式信息系统), Keyword(id=1327013140337897723, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, orderNo=2, keyword=心电图), Keyword(id=1327013140417589500, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, orderNo=3, keyword=量子生成对抗网络), Keyword(id=1327013140501475581, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, orderNo=4, keyword=量子双向门控循环单元), Keyword(id=1327013140606333182, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, orderNo=5, keyword=量子卷积神经网络)], refs=[Reference(id=1327013145689829666, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=Mensah G A, Fuster V, Murray C J L, et al. Global burden of cardiovascular diseases and risks, 1990-2022[J]. Journal of the American College of Cardiology, 2023, 82(25): 2350−2473, articleTitle=null, refAbstract=null), Reference(id=1327013145786298659, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=闫璟辉,宗成庆,徐金安. 中文医疗文本中的嵌套实体识别方法[J]. 软件学报,2024,35(6):2923−2935, articleTitle=null, refAbstract=null), Reference(id=1327013145882767652, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=Yan Jinghui, Zong Chengqing, Xu Jin’an. Configurable text-based image editing by autoencoder-based generative adversarial networks[J]. Journal of Software, 2024, 35(6): 2923−2935 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013146021179685, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=蔡德润,李红燕. 基于度量学习的无监督域适应方法及其在死亡风险预测上的应用[J]. 计算机研究与发展,2022,59(3):674−682, articleTitle=null, refAbstract=null), Reference(id=1327013146096677158, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Cai Derun, Li Hongyan. A metric learning based unsupervised domain adaptation method with its application on mortality prediction[J]. Journal of Computer Research and Development, 2022, 59(3): 674−682 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013146226700583, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=张颖超, 茅丹, 胡凯. 压缩传感理论在心电图信号恢复问题上的研究[J]. 计算机研究与发展,2014,51(5):1018−1027, articleTitle=null, refAbstract=null), Reference(id=1327013146339946792, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Yingchao, Mao Dan, Hu Kai. ECG signal recovery problem based on compressed sensing theory[J]. Journal of Computer Research and Development, 2014, 51(5): 1018−1027 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013146461581609, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[J]. Communications of the ACM, 2020, 63(11): 139−144, articleTitle=null, refAbstract=null), Reference(id=1327013146562244906, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=吴福祥,程俊. 基于自编码器生成对抗网络的可配置文本图像编辑. 软件学报[J],2022,33(9):3139−3151, articleTitle=null, refAbstract=null), Reference(id=1327013146738405675, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Wu Fuxiang, Cheng Jun. Configurable text-based image editing by autoencoder-based generative adversarial networks[J]. Journal of Software, 2022, 33(9): 3139−3151 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013146889400620, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=于海涛,杨小汕,徐常胜. 基于多模态输入的对抗式视频生成方法[J]. 计算机研究与发展,2020,57(7):1522−1530, articleTitle=null, refAbstract=null), Reference(id=1327013147149447469, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Haitao, Yang Xiaoshan, Xu Changsheng. Antagonistic video generation method based on multimodal input[J]. Journal of Computer Research and Development, 2020, 57(7): 1522−1530 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013147245916462, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=蒋斌,刘虹雨,杨超. 一种基于局部属性生成对抗网络的人脸修复算法[J]. 计算机研究与发展,2019,56(11):2485−2493, articleTitle=null, refAbstract=null), Reference(id=1327013147468214575, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Jiang Bin, Liu Hongyu, Yang Chao. A face inpainting algorithm with local attribute generative adversarial networks[J]. Journal of Computer Research and Development, 2019, 56(11): 2485−2493 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013149187879216, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=戴臣超,王洪元,倪彤光. 基于深度卷积生成对抗网络和拓展近邻重排序的行人重识别[J]. 计算机研究与发展,2019,56(8):1632−1641, articleTitle=null, refAbstract=null), Reference(id=1327013149355651377, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Dai Chenchao, Wang Hongyuan, Ni Tongguang. Person reidentification based on deep convolutional generative adversarial network and expanded neighbor reranking[J]. Journal of Computer Research and Development, 2019, 56(8): 1632−1641 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013149435343154, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Brophy E, Wang Zhengwei, She Qi, et al. Generative adversarial networks in time series: A systematic literature review[J]. ACM Computing Surveys, 2023, 55(10): 1−31, articleTitle=null, refAbstract=null), Reference(id=1327013149506646323, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Boixo S, Isakov S V, Smelyanskiy V N, et al. Characterizing quantum supremacy in near-term devices[J]. Nature Physics, 2018, 14(6): 595−600, articleTitle=null, refAbstract=null), Reference(id=1327013149598921012, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Biamonte J, Wittek P, Pancotti N, et al. Quantum machine learning[J]. Nature, 2017, 549(7671): 195−202, articleTitle=null, refAbstract=null), Reference(id=1327013149687001397, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Preskill J. Quantum computing in the NISQ era and beyond[J]. Quantum, 2018, 2: 79−88, articleTitle=null, refAbstract=null), Reference(id=1327013149758304566, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Schuld M, Sinayskiy I, Petruccione F. An introduction to quantum machine learning[J]. Contemporary Physics, 2015, 56(2): 172−185, articleTitle=null, refAbstract=null), Reference(id=1327013149863162167, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Ciliberto C, Herbster M, Ialongo A D, et al. Quantum machine learning: A classical perspective[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2018, 474(2209): 20170551, articleTitle=null, refAbstract=null), Reference(id=1327013149993185593, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=李盼池,周红岩. 基于受控Hadamard门的量子神经网络模型及算法[J]. 计算机研究与发展,2015,52(1):211−220, articleTitle=null, refAbstract=null), Reference(id=1327013150077071674, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Panchi, Zhou Hongyan. Model and algorithm of quantum neural network based on the controlled Hadamard gates[J]. Journal of Computer Research and Development, 2015, 52(1): 211−220 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013150169346363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Benedetti M, Lloyd E, Sack S, et al. Parameterized quantum circuits as machine learning models[J]. Quantum Science and Technology, 2019, 4(4): 043001, articleTitle=null, refAbstract=null), Reference(id=1327013150244843836, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Delaney A M, Brophy E, Ward T E. Synthesis of realistic ECG using generative adversarial networks[J]. arXiv preprint, arXiv: 1909.09150, 2019, articleTitle=null, refAbstract=null), Reference(id=1327013150366478653, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Golany T, Radinsky K. PGANs: Personalized generative adversarial networks for ECG synthesis to improve patient-specific deep ECG classification[C] // Proc of the AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2019: 557−564, articleTitle=null, refAbstract=null), Reference(id=1327013150441976126, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Golany T, Radinsky K, Freedman D. SimGANs: Simulator-based generative adversarial networks for ECG synthesis to improve deep ECG classification[C] // Proc of the Int Conf on Machine Learning. New York: ACM, 2020: 3597−3606, articleTitle=null, refAbstract=null), Reference(id=1327013150563610943, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Vo K, Naeini E K, Naderi A, et al. P2E-WGAN: ECG waveform synthesis from PPG with conditional Wasserstein generative adversarial networks[C] //Proc of the 36th Annual ACM Symp on Applied Computing. New York: ACM, 2021: 1030−1036, articleTitle=null, refAbstract=null), Reference(id=1327013150706217280, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Adib E, Fernandez A S, Afghah F, et al. Synthetic ECG signal generation using probabilistic diffusion models[J]. IEEE Access, 2023, 11: 75818−75828, articleTitle=null, refAbstract=null), Reference(id=1327013150798491969, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Lloyd S, Weedbrook C. Quantum generative adversarial learning[J]. Physical Review Letters, 2018, 121(4): 040502, articleTitle=null, refAbstract=null), Reference(id=1327013150907543874, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Dallaire-Demers P L, Killoran N. Quantum generative adversarial networks[J]. Physical Review A, 2018, 98(1): 012324, articleTitle=null, refAbstract=null), Reference(id=1327013151004012867, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Benedetti M, Garcia-Pintos D, Perdomo O, et al. A generative modeling approach for benchmarking and training shallow quantum circuits[J]. NPJ Quantum Information, 2019, 5(1): 45−56, articleTitle=null, refAbstract=null), Reference(id=1327013151083704644, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Huang Heliang, Du Yuxuan, Gong Ming, et al. Experimental quantum generative adversarial networks for image generation[J]. Physical Review Applied, 2021, 16(2): 024051, articleTitle=null, refAbstract=null), Reference(id=1327013151171785029, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Silver D, Patel T, Cutler W, et al. MosaiQ: Quantum generative adversarial networks for image generation on NISQ computers[C] //Proc of the IEEE/CVF Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 7030−7039, articleTitle=null, refAbstract=null), Reference(id=1327013151251476806, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Qu Zhiguo, Shi Wenke, Tiwari P. Quantum conditional generative adversarial network based on patch method for abnormal electrocardiogram generation[J]. Computers in Biology and Medicine, 2023, 166: 107549, articleTitle=null, refAbstract=null), Reference(id=1327013151343751495, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=韩京宇,陈伟,赵静. 基于异常特征模式的心电数据标签清洗方法[J]. 计算机研究与发展,2023,60(11):2594−2610, articleTitle=null, refAbstract=null), Reference(id=1327013151482163528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=Han Jingyu, Chen Wei, Zhao Jing. A label cleaning method of ECG data based on abnormality-feature patterns[J]. Journal of Computer Research and Development, 2023, 60(11): 2594−2610 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327013151574438217, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=Arjovsky M, Chintala S, Bottou L. Wasserstein generative adversarial networks[C] // Proc of the 34th Int Conf on Machine Learning. New York: PMLR, 2017: 214−223, articleTitle=null, refAbstract=null), Reference(id=1327013151779959114, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=George M. MIT-BIH arrhythmia database [DB/OL]. [2024-05-01]. https://physionet.org/content/mitdb/1.0.0/, articleTitle=null, refAbstract=null), Reference(id=1327013151922565451, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=Paul A. MIT-BIH ST change database [DB/OL]. [2024-05-01]. https://physionet.org/content/stdb/1.0.0/, articleTitle=null, refAbstract=null), Reference(id=1327013151993868620, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=Taddei A. European ST-T database [DB/OL]. [2024-05-01]. https://physionet.org/content/edb/1.0.0/, articleTitle=null, refAbstract=null), Reference(id=1327013153671590222, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=Greenwald S D. Sudden cardiac death holter database [DB/OL]. [2024-05-01]. https://physionet.org/content/sddb/1.0.0/, articleTitle=null, refAbstract=null), Reference(id=1327013153797419343, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=35, rfOrder=43, authorNames=null, journalName=null, refType=null, unstructuredReference=Borji A. Pros and cons of GAN evaluation measures[J]. Computer Vision and Image Understanding, 2019, 179: 41−65, articleTitle=null, refAbstract=null), Reference(id=1327013153893888336, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=36, rfOrder=44, authorNames=null, journalName=null, refType=null, unstructuredReference=Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. arXiv preprint, arXiv: 1511.06434, 2015, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)]), AuthorCompany(id=1327013136298782929, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=2, ext=[AuthorCompanyExt(id=1327013136307171538, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136332337363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学软件工程学院 南京 210044)]), AuthorCompany(id=1327013136437194964, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=3, ext=[AuthorCompanyExt(id=1327013136445583573, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004), AuthorCompanyExt(id=1327013136458166486, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3浙江师范大学计算机科学与技术学院 浙江金华 321004)]), AuthorCompany(id=1327013136537858263, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=4, ext=[AuthorCompanyExt(id=1327013136550441176, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066), AuthorCompanyExt(id=1327013136558829785, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4鹏城实验室新型网络研究部 广东深圳 518066)])], figs=[ArticleFig(id=1327013140799271167, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 1, caption=Overall framework of ECG-QGAN, figureFileSmall=hWDyQtaE8ObC/Mcpa5ANag==, figureFileBig=YNMZW+MD/UmC0+CSgW5KYw==, tableContent=null), ArticleFig(id=1327013140891545856, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图1, caption=ECG-QGAN总体框架, figureFileSmall=hWDyQtaE8ObC/Mcpa5ANag==, figureFileBig=YNMZW+MD/UmC0+CSgW5KYw==, tableContent=null), ArticleFig(id=1327013141109649665, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 2, caption=Illustration of data pretreatment, figureFileSmall=H5yJneBhe+2ZHLbXZD3ntA==, figureFileBig=ohrA3BD7mb7U/gUJO3MzKQ==, tableContent=null), ArticleFig(id=1327013141210312962, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图2, caption=数据预处理示意图, figureFileSmall=H5yJneBhe+2ZHLbXZD3ntA==, figureFileBig=ohrA3BD7mb7U/gUJO3MzKQ==, tableContent=null), ArticleFig(id=1327013141306781955, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 3, caption=Generator structure diagram, figureFileSmall=zUau98I8L9/7AfGD1A7KEw==, figureFileBig=P2x+OzhZMU3gkIMReRpltA==, tableContent=null), ArticleFig(id=1327013141403250948, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图3, caption=生成器结构图, figureFileSmall=zUau98I8L9/7AfGD1A7KEw==, figureFileBig=P2x+OzhZMU3gkIMReRpltA==, tableContent=null), ArticleFig(id=1327013141499719941, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 4, caption=MPS quantum circuit, figureFileSmall=owwBmug3dgP3rskg8HrNdQ==, figureFileBig=DP2jpR9jBRg9DHU3UaaB9Q==, tableContent=null), ArticleFig(id=1327013141587800326, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图4, caption=MPS量子电路图, figureFileSmall=owwBmug3dgP3rskg8HrNdQ==, figureFileBig=DP2jpR9jBRg9DHU3UaaB9Q==, tableContent=null), ArticleFig(id=1327013141722018055, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 5, caption=Discriminator structure, figureFileSmall=eF16J2EJ+czapLi/U8qxhA==, figureFileBig=n89Eac9ChwQtJ1ejUyMnOg==, tableContent=null), ArticleFig(id=1327013141810098440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图5, caption=判别器结构图, figureFileSmall=eF16J2EJ+czapLi/U8qxhA==, figureFileBig=n89Eac9ChwQtJ1ejUyMnOg==, tableContent=null), ArticleFig(id=1327013141902373129, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 6, caption=TTN quantum circuit, figureFileSmall=vH5jo0yH/yMS1R4I660vzw==, figureFileBig=QCcIjfZgyLieseQ0ZYz1FQ==, tableContent=null), ArticleFig(id=1327013141973676298, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图6, caption=TTN量子电路图, figureFileSmall=vH5jo0yH/yMS1R4I660vzw==, figureFileBig=QCcIjfZgyLieseQ0ZYz1FQ==, tableContent=null), ArticleFig(id=1327013142040785163, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 7, caption=Illustration of Dropout, figureFileSmall=ED0veuJ9teuYvKiTOEKHpQ==, figureFileBig=MQFtnu2Z90Z/h7HhzvGvMw==, tableContent=null), ArticleFig(id=1327013142120476940, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图7, caption=Dropout示意图, figureFileSmall=ED0veuJ9teuYvKiTOEKHpQ==, figureFileBig=MQFtnu2Z90Z/h7HhzvGvMw==, tableContent=null), ArticleFig(id=1327013142183391501, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 8, caption=Loss variation graph of generator, figureFileSmall=fnUFFrsvI73ESM1GjTUEMA==, figureFileBig=3OmF/fY0v7vr5c4BJ2pjeQ==, tableContent=null), ArticleFig(id=1327013142267277582, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图8, caption=生成器Loss变化图, figureFileSmall=fnUFFrsvI73ESM1GjTUEMA==, figureFileBig=3OmF/fY0v7vr5c4BJ2pjeQ==, tableContent=null), ArticleFig(id=1327013142346969359, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 9, caption=Discriminator Loss variation diagram for generating ECG data, figureFileSmall=VnOlzs7T+9PKImiNaAQyOA==, figureFileBig=GT7+Yer/0fGzl63piJgBEw==, tableContent=null), ArticleFig(id=1327013142430855440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图9, caption=生成ECG数据的判别器Loss变化图, figureFileSmall=VnOlzs7T+9PKImiNaAQyOA==, figureFileBig=GT7+Yer/0fGzl63piJgBEw==, tableContent=null), ArticleFig(id=1327013142523130129, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 10, caption=Discriminator Loss variation diagram for generating ECG data of class S, figureFileSmall=h5YvYxJh5edSl8Kxinmr0Q==, figureFileBig=lmZNP7JyLdncmr9WRbaLdw==, tableContent=null), ArticleFig(id=1327013142615404818, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图10, caption=生成S类ECG数据的判别器Loss变化图, figureFileSmall=h5YvYxJh5edSl8Kxinmr0Q==, figureFileBig=lmZNP7JyLdncmr9WRbaLdw==, tableContent=null), ArticleFig(id=1327013142695096595, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 11, caption=Comparison of ECG data generated by three models and real ECG data, figureFileSmall=7lgeaPVIq2UtUw1TUkW1nw==, figureFileBig=h0cqSuy9R01XYt9UIoET7w==, tableContent=null), ArticleFig(id=1327013142812537108, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图11, caption=3种模型生成ECG数据与真实ECG数据对比, figureFileSmall=7lgeaPVIq2UtUw1TUkW1nw==, figureFileBig=h0cqSuy9R01XYt9UIoET7w==, tableContent=null), ArticleFig(id=1327013142888034581, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 12, caption=Classification performance of generated ECG data, figureFileSmall=jaKQDmmHTJ4mCxdzscYL5A==, figureFileBig=Uw7fT2uhQbxGchV0m6q3gg==, tableContent=null), ArticleFig(id=1327013142955143446, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图12, caption=生成的ECG数据分类表现, figureFileSmall=jaKQDmmHTJ4mCxdzscYL5A==, figureFileBig=Uw7fT2uhQbxGchV0m6q3gg==, tableContent=null), ArticleFig(id=1327013143030640919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Figure 13, caption=Comparison of theoretical errors and experimental errors under four types of quantum noises, figureFileSmall=/jd0wqDzDdAHvco5ryVJrQ==, figureFileBig=n6Hg84cDvfCrqWtUmDO9ag==, tableContent=null), ArticleFig(id=1327013144704168216, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=图13, caption=在4种量子噪声下的理论错误和实验错误对比, figureFileSmall=/jd0wqDzDdAHvco5ryVJrQ==, figureFileBig=n6Hg84cDvfCrqWtUmDO9ag==, tableContent=null), ArticleFig(id=1327013144825803033, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Table 1, caption=Datasets Used in the Experiment and Their Specific Properties, figureFileSmall=null, figureFileBig=null, tableContent=
数据集频率/Hz记录数时间/h标记
文献[31]360480.5全部
文献[32]36028不固定全部
文献[33]250902全部
文献[34]250237~25部分
), ArticleFig(id=1327013144901300506, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=表1, caption=

实验中使用的数据集以及其具体属性

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集频率/Hz记录数时间/h标记
文献[31]360480.5全部
文献[32]36028不固定全部
文献[33]250902全部
文献[34]250237~25部分
), ArticleFig(id=1327013145035518235, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Table 2, caption=Comparison of ECG Signals Generated by Models on PCC, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN0.13460.13200.06470.03780.4389
QPatch-GAN0.15310.10730.13860.03120.5438
QCGAN-ECG0.17440.17400.15150.03420.6323
BiGRU-CNN GAN0.15690.17980.13480.04360.5596
BiLSTM-CNN GAN0.16590.15140.14900.04070.5850
ECG-QGAN(本文)0.21870.23390.15240.05160.6067
), ArticleFig(id=1327013145115210012, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=表2, caption=

模型生成ECG信号在PCC上对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN0.13460.13200.06470.03780.4389
QPatch-GAN0.15310.10730.13860.03120.5438
QCGAN-ECG0.17440.17400.15150.03420.6323
BiGRU-CNN GAN0.15690.17980.13480.04360.5596
BiLSTM-CNN GAN0.16590.15140.14900.04070.5850
ECG-QGAN(本文)0.21870.23390.15240.05160.6067
), ArticleFig(id=1327013145253622046, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Table 4, caption=Comparison of ECG Signals Generated by Models on RMSE, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN0.57250.35220.79040.56430.4857
QPatch-GAN0.58020.32820.77380.81990.5279
QCGAN-ECG0.56230.30670.71230.75030.4921
BiGRU-CNN GAN0.29970.12600.71320.50510.4431
BiLSTM-CNN GAN0.19330.11890.72500.50430.4422
ECG-QGAN(本文)0.37630.14590.71540.50140.4418
), ArticleFig(id=1327013145341702431, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=表4, caption=

模型生成ECG信号在RMSE上对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN0.57250.35220.79040.56430.4857
QPatch-GAN0.58020.32820.77380.81990.5279
QCGAN-ECG0.56230.30670.71230.75030.4921
BiGRU-CNN GAN0.29970.12600.71320.50510.4431
BiLSTM-CNN GAN0.19330.11890.72500.50430.4422
ECG-QGAN(本文)0.37630.14590.71540.50140.4418
), ArticleFig(id=1327013145463337248, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=EN, label=Table 3, caption=Comparison of ECG Signals Generated by Models on PRD, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN1.8021.5141.4801.7791.508
QPatch-GAN1.0771.1341.3541.1830.873
QCGAN-ECG1.0611.1281.4441.1160.896
BiGRU-CNN GAN1.2361.0161.1731.1290.830
BiLSTM-CNN GAN1.0950.9921.1311.0540.829
ECG-QGAN(本文)0.9841.0181.0851.0550.752
), ArticleFig(id=1327013145538834721, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, language=CN, label=表3, caption=

模型生成ECG信号在PRD上对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型N类S类V类F类Q类
注:黑体表示最优值.
DCGAN1.8021.5141.4801.7791.508
QPatch-GAN1.0771.1341.3541.1830.873
QCGAN-ECG1.0611.1281.4441.1160.896
BiGRU-CNN GAN1.2361.0161.1731.1290.830
BiLSTM-CNN GAN1.0950.9921.1311.0540.829
ECG-QGAN(本文)0.9841.0181.0851.0550.752
)], attaches=null, journal=Journal(id=1324759606947315908, delFlag=0, nameCn=计算机研究与发展, nameEn=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, issn=1000-1239, eissn=null, cn=11-1777/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=k2l8zgqwstuxxevFX2qxVA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Computer Research and Development, journalRemark=null, publicationField=null, createdTime=1793776903141, updatedTime=1793850698612, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=k2l8zgqwstuxxevFX2qxVA==, picEn=r6k0Ry+RD+c/VXj5QarmDw==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069127710286688, language=CN, name=计算机研究与发展, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698640, updatedTime=1793850698640, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069127777395553, language=EN, name=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698656, updatedTime=1793850698656, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759914146529478, websiteList=[Website(id=1325070300139892771, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN, language=CN, createTime=1793850978169, createBy=18614031015, updateTime=1793851040259, updateBy=18614031015, name=计算机研究与发展-中文, tplId=1146099689490845704, title=计算机研究与发展, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070893751349360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=articleTextType, value=kx, createTime=1793851119697, updateTime=1793851119697, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893730377837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=banner, value=null, createTime=1793851119692, updateTime=1793851119692, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893772320883, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=grayFlag, value=0, createTime=1793851119702, updateTime=1793851119702, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893721989228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851119690, updateTime=1793851119690, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893784903797, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=minRunFlag, value=0, createTime=1793851119705, updateTime=1793851119705, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893742960751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic, createTime=1793851119695, updateTime=1793851119695, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893780709492, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=silenceFlag, value=0, createTime=1793851119704, updateTime=1793851119704, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893738766446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851119694, updateTime=1793851119694, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893755543665, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeColor, value=null, createTime=1793851119698, updateTime=1793851119698, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893763932274, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeStyle, value=null, createTime=1793851119700, updateTime=1793851119700, creator=18614031015, updator=18614031015)]), Website(id=1325070300580294698, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN, language=EN, createTime=1793850978274, createBy=18614031015, updateTime=1793851036638, updateBy=18614031015, name=计算机研究与发展-英文, tplId=1146101810881728533, title=Journal of Computer Research and Development, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070925288321148, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=articleTextType, value=kx, createTime=1793851127216, updateTime=1793851127216, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925263155321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=banner, value=null, createTime=1793851127210, updateTime=1793851127210, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925309292671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=grayFlag, value=0, createTime=1793851127221, updateTime=1793851127221, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925254766712, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851127208, updateTime=1793851127208, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925326069889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=minRunFlag, value=0, createTime=1793851127225, updateTime=1793851127225, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925279932539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic, createTime=1793851127214, updateTime=1793851127214, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925317681280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=silenceFlag, value=0, createTime=1793851127223, updateTime=1793851127223, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925271543930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851127212, updateTime=1793851127212, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925292515453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeColor, value=null, createTime=1793851127217, updateTime=1793851127217, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925300904062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeStyle, value=null, createTime=1793851127219, updateTime=1793851127219, creator=18614031015, updator=18614031015)])], journalTitle=计算机研究与发展, weixinUrl=null, journalUrl=https://crad.ict.ac.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Computer Research and Development, journalPhotoCn=k2l8zgqwstuxxevFX2qxVA==, journalPhotoEn=r6k0Ry+RD+c/VXj5QarmDw==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202440527, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202440527, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202440527, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202440527, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1751299200000, fullTextJson=null, articleText=null, reference=null)
收藏切换
ECG-QGAN:基于量子生成对抗网络的心电图生成式信息系统
瞿治国 1 , 陈韦龙 1 , 孙乐 1, * , 刘文杰 2 , 张彦春 3, 4
计算机研究与发展 | 生成式人工智能驱动的信息系统专题 2025,62(7): 1622-1638
收起
收藏切换
计算机研究与发展 |生成式人工智能驱动的信息系统专题 2025 , 62 (7) : 1622 -1638
ECG-QGAN:基于量子生成对抗网络的心电图生成式信息系统
全屏
[Author(id=1327013136634327259, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013136734990557, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136634327259, language=EN, stringName=Zhiguo Qu, firstName=Zhiguo, middleName=null, lastName=Qu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

"}, bioImg=null, bioContent=

born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013136839848158, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136634327259, language=CN, stringName=瞿治国, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"z3C66fWIRK5JM6qRro/fIg==","content":"

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

"}, bioImg=z3C66fWIRK5JM6qRro/fIg==, bioContent=

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013136940511456, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137028591842, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136940511456, language=EN, stringName=Weilong Chen, firstName=Weilong, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 2000. Master. His main research interest includes quantum neural networks

"}, bioImg=null, bioContent=

born in 2000. Master. His main research interest includes quantum neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137137643748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013136940511456, language=CN, stringName=陈韦龙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"tscjdL/Opx2ZOLF+bzNYGQ==","content":"

2000年生. 硕士. 主要研究方向为量子神经网络

"}, bioImg=tscjdL/Opx2ZOLF+bzNYGQ==, bioContent=

2000年生. 硕士. 主要研究方向为量子神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013137229918438, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=002359@nuist.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327013137317998824, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137229918438, language=EN, stringName=Le Sun, firstName=Le, middleName=null, lastName=Sun, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

"}, bioImg=null, bioContent=

born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137406079209, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137229918438, language=CN, stringName=孙乐, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044, bio={"img":"fQSakqWUzx13IgoxVJhigw==","content":"

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

"}, bioImg=fQSakqWUzx13IgoxVJhigw==, bioContent=

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136214896846, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=1, ext=[AuthorCompanyExt(id=1327013136227479759, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136235868368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136214896846, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京信息工程大学计算机学院、网络空间安全学院 南京 210044)])]), Author(id=1327013137502548203, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137582239981, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137502548203, language=EN, stringName=Wenjie Liu, firstName=Wenjie, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044, bio={"content":"

born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137670320366, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137502548203, language=CN, stringName=刘文杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2南京信息工程大学软件工程学院 南京 210044, bio={"img":"qvghsECXHdxeTrw9TzF3sg==","content":"

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

"}, bioImg=qvghsECXHdxeTrw9TzF3sg==, bioContent=

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136298782929, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=2, ext=[AuthorCompanyExt(id=1327013136307171538, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044), AuthorCompanyExt(id=1327013136332337363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136298782929, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2南京信息工程大学软件工程学院 南京 210044)])]), Author(id=1327013137787760880, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327013137917784307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137787760880, language=EN, stringName=Yanchun Zhang, firstName=Yanchun, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004
4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066, bio={"content":"

born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

"}, bioImg=null, bioContent=

born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327013137997476084, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, authorId=1327013137787760880, language=CN, stringName=张彦春, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, 4, address=3浙江师范大学计算机科学与技术学院 浙江金华 321004
4鹏城实验室新型网络研究部 广东深圳 518066, bio={"img":"IOt9r0FLojX/7aQzdnXNxg==","content":"

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

"}, bioImg=IOt9r0FLojX/7aQzdnXNxg==, bioContent=

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327013136437194964, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=3, ext=[AuthorCompanyExt(id=1327013136445583573, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004), AuthorCompanyExt(id=1327013136458166486, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136437194964, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3浙江师范大学计算机科学与技术学院 浙江金华 321004)]), AuthorCompany(id=1327013136537858263, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, xref=4, ext=[AuthorCompanyExt(id=1327013136550441176, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066), AuthorCompanyExt(id=1327013136558829785, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327013128828727450, companyId=1327013136537858263, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4鹏城实验室新型网络研究部 广东深圳 518066)])])]
瞿治国1, 陈韦龙1, 孙乐1, * , 刘文杰2, 张彦春3, 4
作者信息
  • 1南京信息工程大学计算机学院、网络空间安全学院 南京 210044
  • 2南京信息工程大学软件工程学院 南京 210044
  • 3浙江师范大学计算机科学与技术学院 浙江金华 321004
  • 4鹏城实验室新型网络研究部 广东深圳 518066
通讯作者:
作者简介:

1976年生. 博士,副教授. CCF会员. 主要研究方向为量子机器学习、深度学习、区块链和量子区块链、量子安全通信

2000年生. 硕士. 主要研究方向为量子神经网络

1987年生. 博士,教授. CCF会员. 主要研究方向为图神经网络、节能智能医疗、数据挖掘

1979年生. 博士,副教授. CCF会员. 主要研究方向为量子安全多方计算、量子密码分析、量子机器学习、知识图谱与图神经网络、大模型

1958年生. 博士,教授. 主要研究方向为数据库、数据挖掘、健康信息学、网络信息系统、网络服务

ECG-QGAN: An ECG Generative Information System Based on Quantum Generative Adversarial Networks
Zhiguo Qu1, Weilong Chen1, Le Sun1, * , Wenjie Liu2, Yanchun Zhang3, 4
Affiliations
  • 1School of Computer Science & School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing 210044
  • 2School of Software, Nanjing University of Information Science and Technology, Nanjing 210044
  • 3School of Computer Science and Technology, Zhejiang Normal University, Jinhua, Zhejiang 321004
  • 4Department of New Networks, Peng Cheng Laboratory, Shenzhen, Guangdong 518066
  • born in 1976. PhD, associate professor. Member of CCF. His main research interests include quantum machine learning, deep learning, blockchain and quantum blockchain, and quantum secure communication

    born in 2000. Master. His main research interest includes quantum neural networks

    born in 1987. PhD, professor. Member of CCF. Her main research interests include graph neural networks, energy-efficient smart healthcare, and data mining

    born in 1979. PhD, associate professor. Member of CCF. His main research interests include quantum secure multi-party computation, quantum cryptanalysis, quantum machine learning, knowledge graph and graph neural network, and large model

    born in 1958. PhD, professor. His main research interests include database, data mining, health informatics, Web information systems, and Web services

出版时间: 2025-07-01 doi: 10.7544/issn1000-1239.202440527
文章导航
收藏切换

据统计,我国心血管疾病患病人数约达3.3亿,每年因为心血管疾病死亡的人数占总死亡人数的40%. 在这种背景下,心脏病辅助诊断系统的发展显得尤为重要,但其开发受限于缺乏不含患者隐私信息和由医疗专家标注的大量心电图(electrocardiogram, ECG)临床数据. 作为一门新兴学科,量子计算可通过利用量子叠加和纠缠特性,能够探索更大、更复杂的状态空间,进而有利于生成同临床数据一样的高质量和多样化的ECG数据. 为此,提出了一种基于量子生成对抗网络(QGAN)的ECG生成式信息系统,简称ECG-QGAN. 其中QGAN由量子双向门控循环单元(quantum bidirectional gated recurrent unit, QBiGRU)和量子卷积神经网络(quantum convolutional neural network, QCNN)组成. 该系统利用量子的纠缠特性提高生成能力,以生成与现有临床数据一致的ECG数据,从而可以保留心脏病患者的心跳特征. 该系统的生成器和判别器分别采用QBiGRU和 QCNN,并应用了基于矩阵乘积状态(matrix product state, MPS)和树形张量网络(tree tensor network, TTN)所设计的变分量子电路(variational quantum circuit , VQC),可以使该系统在较少的量子资源下更高效地捕捉ECG数据信息,生成合格的ECG数据. 此外,该系统应用了量子Dropout技术,以避免训练过程中出现过拟合问题. 最后,实验结果表明,与其他生成ECG数据的模型相比,ECG-QGAN生成的ECG数据具有更高的平均分类准确率. 同时它在量子位数量和电路深度方面对当前噪声较大的中尺度量子(noise intermediate scale quantum, NISQ)计算机是友好的.

生成式信息系统  /  心电图  /  量子生成对抗网络  /  量子双向门控循环单元  /  量子卷积神经网络

According to statistics, the number of people suffering from cardiovascular diseases in China is about 330 million, and the number of deaths caused by cardiovascular diseases accounts for 40% of the total number of deaths each year. Under this circumstance, the development of heart disease assisted diagnosis systems is particularly important, but its development is limited by the lack of a large amount of electrocardiogram (ECG) clinical data that don’t contain patient privacy information and need to be annotated by medical experts. As an emerging discipline, quantum computing can explore larger and more complex state spaces by utilizing quantum superposition and entanglement properties, which is beneficial for generating high-quality and diverse electrocardiogram data similar to real clinical data. Therefore, we propose an electrocardiogram generative information system based on quantum generative adversarial networks, abbreviated as ECG-QGAN. The quantum generative adversarial network consists of a quantum bidirectional gated recurrent unit (QBiGRU) and a quantum convolutional neural network (QCNN). The system utilizes the entanglement property of quantum to improve the generative capability to produce ECG data that are consistent with the existing clinical data so that the heartbeat characteristics of cardiac patients can be preserved. The generator and discriminator of this system use QBiGRU and QCNN, respectively. The variational quantum circuit (VQC) designed based on matrix product state (MPS) and tree tensor network (TTN) is adopted, which enables the system to capture ECG data information more efficiently and generate qualified ECG data with fewer quantum resources. In addition, the system applies quantum Dropout technology to avoid overfitting issues during the training process. Finally, the experimental results show that the ECGs generated by ECG-QGAN have a higher average classification accuracy compared with other models for generating ECGs. It is also friendly to the current noise intermediate scale quantum (NISQ) computers in terms of the number of quantum bits and circuit depth.

generative information system  /  electrocardiogram (ECG)  /  quantum generative adversarial network (QGAN)  /  quantum bidirectional gated recurrent unit (QBiGRU)  /  quantum convolutional neural network (QCNN)
瞿治国, 陈韦龙, 孙乐, 刘文杰, 张彦春. ECG-QGAN:基于量子生成对抗网络的心电图生成式信息系统. 计算机研究与发展, 2025 , 62 (7) : 1622 -1638 . DOI: 10.7544/issn1000-1239.202440527
Zhiguo Qu, Weilong Chen, Le Sun, Wenjie Liu, Yanchun Zhang. ECG-QGAN: An ECG Generative Information System Based on Quantum Generative Adversarial Networks[J]. Journal of Computer Research and Development, 2025 , 62 (7) : 1622 -1638 . DOI: 10.7544/issn1000-1239.202440527
心血管疾病是导致人类死亡的主要疾病,其中大部分是由心脏病所引起. 因此,各国每年都需要投入大量医疗资源用于心血管疾病的诊断[1]. 心电图(electrocardiogram, ECG)是一种非平稳的生理信号,代表心脏的电活动. 它含有丰富的病理信息,是诊断心血管疾病的重要临床工具. 每次心跳都可以通过由5个峰值(特征)组成的心电图波形来准确描述. 而根据心电图每个峰值的不同特征,人们可以对心跳节拍进行分类. 美国医疗器械促进协会(The Association for the Advancement of Medical Instrumentation,AAMI)将心律失常心跳细分为5类:室性异位搏动(V)、不可分类的节拍(Q)、正常心搏(N)、融合搏动(F)和室上性异位搏动(S). 传统的心血管疾病诊断手段,依赖于患者的病史、临床检查以及一系列医学参数来阐释和划分疾病类型. 然而,这种方法在面对海量的异构数据时,往往效率低下,同时也消耗了大量的医疗资源. 传统诊断方法的局限性促使对可靠、自动且低成本的监测和诊断系统的需求.
深度神经网络(deep neural network,DNN)已广泛用于不同领域的分类和预测任务,这对各种医疗任务产生了显著影响[2-3]. 近些年来,多种基于DNN的ECG智能辅助诊断系统被相继提出. 这些系统可以有效地克服传统心血管疾病诊断的局限,辅助医生诊断以及实时监测患者心跳. 然而,这些ECG智能辅助诊断系统面临的主要挑战之一是需要大量的ECG数据进行训练. 在医学研究中,尤其是ECG数据,通常会受到数据稀缺、数据标注成本和数据不平衡的困扰[4]. 同时真实的ECG数据会涉及到患者隐私问题. 因此,寻找合适的生成式信息系统以扩充ECG训练数据集是十分必要. 生成对抗网络(generative adversarial network, GAN)[5]自诞生以来,在深度学习领域引起了广泛的关注,它们能够生成和处理跨多个领域的高质量数据. 近年来,GAN 已展示出在高质量图像和视频生成、风格迁移以及图像补全方面的卓越能力[6-9]. 它们也已成功应用于音频生成、序列预测和插补,并逐渐发展为对时间序列和顺序数据的生成和预测[10]. 作为时间序列数据的一种,ECG数据通过GAN的应用可以解决当前ECG智能辅助诊断系统所面临的有效数据缺少的问题. 但由于GAN的训练是一个对抗博弈过程,在处理问题时对计算速度与计算效率有更高的要求. 而经典神经网络受限于串行计算模式,在计算速度与计算效率方面的提升是一个挑战. 同时,GAN在处理高维数据时同样面临挑战,因为高维数据的复杂性和多样性使得生成器难以捕捉到所有有用的特征,这将导致生成样本的质量会出现问题. 因此,寻找更高效、更强大的计算模型成为了当务之急.
近几十年来,作为新兴计算模型的量子计算一直受到广泛关注[11]. 由于量子力学可以生成反直觉数据,因此人们相信量子计算机可以识别出经典计算机难以识别的数据[12]. 随着噪声较大的中尺度量子(noise intermediate scale quantum,NISQ)计算机的发展[13],研究如何有效地利用有限的量子资源寻找最有前途的应用成为了关键问题. NISQ指的是目前可以实现的量子计算机的一个阶段,这些量子计算机具有几十到几百个量子比特,可以达到中等规模. 尽管量子比特仍然受到噪声和误差的影响,但NISQ计算机已经可以在某些特定的领域和问题上展示出潜在的优势.NISQ被认为是从当前的量子计算技术向未来大规模容错量子计算机过渡的一个重要阶段,将为最终实现实用的量子计算机打下基础. 机器学习是量子计算机的一个非常有吸引力的应用领域. 由于机器学习和量子力学都基于统计学原理,且机器学习技术本质上具有对抗噪声的能力,它可能促使近期量子计算机无需纠错即可运行[14]. 量子机器学习将机器学习和量子计算相结合,不仅能改进量子计算系统的基准测试、理解和控制,还有望显著增强一些机器学习算法,如减少计算复杂性、加快训练速度和提升模型性能等方面[15-16]. 量子生成对抗网络是一种结合量子计算和经典GAN的量子机器学习算法,旨在推动生成学习. 在数据生成方面,量子优势在于可以生成经典的神经网络无法生成的数据,这意味着量子GAN在生成任务上具有很大的研究价值. 量子GAN生成的ECG数据可与真实数据一同用于训练心脏病辅助诊断系统. 这些生成的数据自动带有标签,不需要拥有专业知识的医疗专家进行标注,从而降低标注成本. 此外,这些生成数据不会涉及患者隐私,可广泛地用于训练. 因此,构建以量子生成对抗网络(quantum generative adversarial network,QGAN)为主体的生成式信息系统,是扩充ECG数据集的有效方法,是对心脏病辅助诊断系统研究和发展的重要保障. 虽然当前量子硬件的资源限制使得执行大规模量子算法变得困难,但是许多量子机器学习研究如何与 NISQ 设备兼容的量子算法,如开发具有参数化量子电路 (parameter quantum circuit,PQC)或变分量子电路(variational quantum circuit,VQC)的混合量子经典方案[17]. 出于以上考虑,本文提出了ECG-QGAN,一种基于量子生成对抗网络的心电图生成式信息系统. 其中生成器为量子双向门控循环单元(quantum bidirectional gated recurrent unit,QBiGRU),判别器为量子卷积神经网络(quantum convolutional neural network,QCNN). 本文的主要贡献包括3个方面:
1)提出一种基于QGAN的生成式信息系统,用于解决ECG生成问题. 生成器采用QBiGRU,其目的是利用量子计算能力和纠缠特性以及BiGRU处理时间序列数据的优势,提高对ECG数据的生成能力;判别器采用QCNN,其目的是利用量子特性和CNN强大的特征提取能力,有效区分真实数据和生成数据.
2)为了与NISQ设备兼容,本文还分别为QBiGRU和QCNN设计基于矩阵乘积状态(matrix product state,MPS)和树状张量网络(tree tensor network,TTN)的变分量子电路,可以更好地捕捉到ECG数据的特征,同时有效减少所需的参数数量,进一步提高量子电路的稳定性和可靠性.
3)为了避免量子神经网络出现的过拟合问题,本文还引入了量子Dropout,使ECG-QGAN达到更好的泛化效果.
ECG对诊断心血管疾病至关重要. 利用机器学习诊断心血管疾病可以显著提高临床医生的工作效率. 然而,由于ECG属于医疗数据,其使用存在一定局限性:数据量有限、类别间数据量不均衡以及需保护患者隐私. 这些限制妨碍了其在临床研究中的应用. 因此,研究不受这些限制的生成式ECG数据引起越来越多的关注.
2019年,Delaney等人[18]开发了一系列 GAN 架构来生成合成正弦波和合成心电图. 然后使用2个评估指标来定量评估合成数据在临床培训和数据分析等实际应用中的适用性. 并从定量和定性2方面证明,GAN 架构可以成功生成时间序列信号. 这些信号不仅在结构上与训练集相似,而且在生成的样本中具有多样性. 最后还报告了它们抵御简单成员推断攻击的能力,从而保护了训练集的隐私. 同年,Golany等人[19]提出了一种称为PGANs(personalized generative adversarial networks )生成模型,该模型可学习合成患者特异性心电信号,然后将其用作额外的训练数据,以提高患者特异性分类器的性能. 2020年,Golany等人[20]又提出一种模型SimGANs(simulator-based generative adversarial networks),该模型研究了如何利用生物模拟器将生物生理的动态过程纳入ECG分类任务的生成过程. 具体来说,就是使用代表心脏动力学的常微分方程系统,并将此常微分方程系统纳入GAN的优化过程,以创建生物学上可信的ECG训练示例. 2021年,Vo等人[21]提出了P2E-WGAN:一种纯端到端、可泛化的深度学习模型,使用条件瓦瑟斯坦GAN从PPG合成ECG波形. 他们提出的生成模型能够增加训练数据,缓解机器学习方法的数据饥渴问题. 2023年, Adib等人[22]对比了改进型去噪扩散概率模型(DDPM)和带梯度惩罚的WGAN-GP(Wasserstein GAN)模型生成合成心电信号的能力. 通过对每个模型生成的心电信号的质量、分布和真实性进行了定量评估和比较,得出WGAN-GP 模型在所有考虑的指标中始终优于 DDPM.
近年来,量子计算具有处理指数级别的高维数据的卓越能力,这使得量子机器学习成为一个热门研究方向. 因此,自然要考虑如何利用量子计算来改进其生成模型. 实际上,已经有若干努力致力于这一问题.
2018年,Lloyd等人[23]引入了QGAN的概念,其中的数据由量子态或经典数据组成,生成器和判别器都配备了量子信息处理器. 他们证明,当生成器产生与数据相同的统计量时,量子对抗博弈的唯一定点也会出现(量子对抗博弈是将量子计算的原理应用于博弈论的一种研究领域. 在这种博弈中,玩家使用量子状态和量子策略进行对抗. 这种博弈利用量子叠加和量子纠缠的特性,使得玩家的策略和结果比经典博弈更加复杂和多样. 量子对抗博弈展示了量子计算在对抗性环境中的独特优势). 并且当数据由高维空间上的测量样本组成时,量子对抗网络与经典对抗网络相比,可能会表现出指数级的加速计算优势. 同年,Dallaire-Demers等人[24]将对抗训练扩展到量子领域,并展示了如何利用量子电路构建GAN. 他们还展示了如何利用另一种量子电路计算梯度这一GAN训练的关键要素. 他们设计了一个简单实用的电路,用于参数化量子机器学习模型,并进行了一个简单的数值实验,证明QGAN可以成功训练. 2019年,Benedetti等人[25]为了扩展混合量子经典算法技术的组合,提出了一种量子电路学习算法,可用于协助量子设备的表征,并训练浅层电路完成生成任务. 该程序通过使用原生门及其量子位连接,最大程度地利用了量子硬件能力. 经过证明,这种算法可以学习格林伯格-霍恩-蔡林格态的最佳制备方法(格林伯格-霍恩-蔡林格态是量子力学中的一种特殊多粒子纠缠态,这种态因物理学家丹尼尔·格林伯格、迈克尔·霍恩和安东·蔡林格的研究而得名,它能更深刻地揭示量子纠缠的本质,并否定了局域隐变量理论,从而进一步验证了量子力学的完备性). 2021年,Huang等人[26]设计了一种灵活的QGAN方案. 原则上,该方案能够完成具有高维特征的图像生成,并能利用量子叠加来并行训练多个示例. 他们通过实验在超导量子处理器上实现了真实世界手写数字图像的学习和生成. 他们为在近期量子设备上开发先进的量子生成模型提供了指导,并为探索量子在各种 GAN 相关学习任务中的优势开辟了一条途径. 2023年,Silver等人[27]引入了MosaiQ,这是一个高质量的量子图像生成 GAN 框架,可以在现有的NISQ计算机上执行,并且解决了以往的技术在质量和鲁棒性方面存在的不足. 同年,Qu等人[28]提出了一种用于生成异常心电信号的新型量子条件生成对抗算法QCGAN-ECG. QCGAN-ECG 采用补丁法构建量子生成器,节省了量子资源,使 QCGAN-ECG 成为近期量子设备的实用工具. 此外,QCGAN-ECG 还引入了量子寄存器作为控制条件. 它将异常心跳的类型和概率分布信息编码到量子寄存器中,使整个生成过程可控. QCGAN-ECG 还能准确拟合各种异常ECG数据的概率分布. 在抗噪声实验中,QCGAN-ECG 在不同程度的量子噪声干扰下表现出卓越的鲁棒性.
虽然近年来有不少QGAN算法被陆续提出,但其仍处于发展初期. 同时,只有极少数的QGAN算法能用于生成ECG信号. 因此,QGAN在ECG生成方面的潜力值得我们进一步深入挖掘.
量子计算是一种基于量子力学原理的计算方式. 量子比特是量子计算的核心,量子计算与量子信息建立在量子比特的基础上. 经典比特有0和1这2个状态,同样的量子比特也有$\left| 0 \right\rangle $和$\left| 1 \right\rangle $两个状态,记号$ “|\rangle ” $称为Dirac记号,在量子力学中用于表示量子状态. 量子比特可以是状态的线性组合,常称为叠加态,这一特性使得量子计算机在执行计算任务时,能够同时考虑多种可能的情况,大幅提高计算效率. 叠加态如式(1)所示:
$ \left| {\boldsymbol{\psi}} \right\rangle = \alpha \left| 0 \right\rangle + \beta \left| 1 \right\rangle \text{,} $
其中$\alpha $和$\beta $均是复数,满足$ {\left| \alpha \right|^2} + {\left| \beta \right|^2} = 1 $. 量子比特的状态是2维复向量空间中的向量,特殊的$\left| 0 \right\rangle $和$\left| 1 \right\rangle $状态称为计算基态,是构成这个向量空间的一组正交基. 同时,因为$ {\left| \alpha \right|^2} + {\left| \beta \right|^2} = 1 $,式(1)可以改写为
$ \left| {\boldsymbol{\psi}} \right\rangle = {{\mathrm{e}}^{{\mathrm{i}}\gamma }}\left(\cos \frac{\theta }{2}\left| 0 \right\rangle + {{\mathrm{e}}^{{\mathrm{i}}\tau }}\sin \frac{\theta }{2}\left| 1 \right\rangle \right), $
其中$\theta $,$\gamma $,$\tau $都是实数,由于${{\mathrm{e}}^{{\mathrm{i}}\gamma }}$不具有任何可观测的效应,可以略去,因此式(2)可以改写为有效的形式
$ \left| {\boldsymbol{\psi}} \right\rangle = \left(\cos \frac{\theta }{2}\left| 0 \right\rangle + {{\mathrm{e}}^{{\mathrm{i}}\tau }}\sin \frac{\theta }{2}\left| 1 \right\rangle \right). $
量子纠缠是指2个或多个粒子在量子状态上相互关联,导致其中一个粒子的状态改变会立即影响到其他粒子的状态,即使它们相距很远. 这种现象是由量子力学中的叠加原理和不可分割性引起的. 具体来说,当2个粒子纠缠在一起时,它们的量子态不能独立描述,而是必须用一个联合的量子态来描述. 这意味着测量一个粒子的状态会立即确定另一个粒子的状态. 纠缠粒子之间的这种强关联性使得它们在信息传递和计算中具有独特的优势. 密度算子(density operator)在描述未知的量子系统和复合量子系统时更具优越性. 设一个量子系统以概率${p_i}$处于状态$\left| {{{\boldsymbol{\varphi}} _i}} \right\rangle $,则称$\left\{ {{p_i}\left| {{{\boldsymbol{\varphi}} _i}} \right\rangle } \right\}$为一个系综,且满足$\sum\limits_i {p_i} = 1$. 该系统的密度算子表示为
$ {\boldsymbol{\rho}} = \sum\limits_i {{p_i}\left| {{{\boldsymbol{\varphi}} _i}} \right\rangle } \left\langle {{{\boldsymbol{\varphi}} _i}} \right| .$
如果某个系统以概率${p_i}$处于状态${{\boldsymbol{\rho}} _i}$,则系统的密度算子表示为
$ {\boldsymbol{\rho}} = \sum\limits_i {{p_i}{{\boldsymbol{\rho}} _i}} . $
若2个子系统分别处于状态${{\boldsymbol{\rho}} _1}$和${{\boldsymbol{\rho}} _2}$,则其复合系统可由${{\boldsymbol{\rho}} _1} \otimes {{\boldsymbol{\rho}} _{2}}$表示,其中$ \otimes $表示张量积.
量子门是对量子比特进行操作的基本单元,类似于经典计算中的逻辑门. 常见的单量子门有Hadamard门(使量子比特置于叠加态)、${{\boldsymbol{R}}_x}$旋转门(使量子比特在Bloch球面上绕x轴旋转)和${{\boldsymbol{R}}_y}$旋转门(使量子比特在Bloch球面上绕y轴旋转),如式(6)~(8)所示. 多量子比特门有CNOT门(对2个量子比特操作,第2个量子比特的状态由第1个量子比特的状态控制),如式(9)所示.
$ {\boldsymbol{H}} = \frac{1}{{\sqrt 2 }}\left( {\begin{array}{*{20}{c}}1&1 \\ 1&{ - 1} \end{array}} \right) , $
$ {{\boldsymbol{R}}_x}\left( \theta \right) = \left( {\begin{array}{*{20}{c}}{\cos \dfrac{\theta }{2}}&{ - {\mathrm{i}}\sin \dfrac{\theta }{2}} \\ { - {\mathrm{i}}\sin \dfrac{\theta }{2}}&{\cos \dfrac{\theta }{2}} \end{array}} \right), $
$ {{\boldsymbol{R}}_y}(\theta ) = \left( {\begin{array}{*{20}{c}}{{{\mathrm{e}}^{ - {\mathrm{i}}\theta /2}}}&0 \\ 0&{{{\mathrm{e}}^{{\mathrm{i}} \theta /2}}} \end{array}} \right), $
$ {\boldsymbol{CNOT}} = \left( {\begin{array}{*{20}{c}}{\begin{array}{*{20}{c}}1&0 \\ 0&1 \end{array}}&{\begin{array}{*{20}{c}}0&0 \\ 0&0 \end{array}} \\ {\begin{array}{*{20}{c}}0&0 \\ 0&0 \end{array}}&{\begin{array}{*{20}{c}}0&1 \\ 1&0 \end{array}} \end{array}} \right) . $
通用量子门是量子计算中的基本组成部分,通过合理的设计和组合这些量子门,可以实现对量子系统的任意操作. 一个初态为$\left| {\boldsymbol{\psi}} \right\rangle $经酉矩阵U演化为$\left| {{\boldsymbol{\psi}} '} \right\rangle $可以表示为
$ \left| {{\boldsymbol{\psi}} '} \right\rangle = {\boldsymbol{U}}\left| {\boldsymbol{\psi}} \right\rangle . $
量子测量通过一组作用于被测系统状态空间的测量算子$\left\{ {{{\boldsymbol{M}}_m}} \right\}$描述,其中m表示实验中可能的测量结果. 这些算子在量子信息提取中起着关键作用. 若测量前量子系统的最新状态已知是$\left| {\boldsymbol{\varphi}} \right\rangle $,那么得到结果m的概率可以计算得出
$ p(m) = \left\langle {\boldsymbol{\varphi}} \right|{\boldsymbol{M}}_m^\dagger {{\boldsymbol{M}}_m}\left| {\boldsymbol{\varphi}} \right\rangle , $
且测量后的系统状态坍缩为
$ \frac{{{{\boldsymbol{M}}_m}\left| {\boldsymbol{\varphi}} \right\rangle }}{{\sqrt {\left\langle {\boldsymbol{\varphi}} \right|{\boldsymbol{M}}_m^\dagger {{\boldsymbol{M}}_m}\left| {\boldsymbol{\varphi}} \right\rangle } }} , $
其中测量算子$\left\{ {{{\boldsymbol{M}}_m}} \right\}$必须满足完备性方程:
$ \sum\limits_m {{\boldsymbol{M}}_m^\dagger {{\boldsymbol{M}}_m} = {\boldsymbol{I}}}. $
由完备性方程可以推算得知不同测量结果的概率和为
$ \sum\limits_m {p(m) = \sum\limits_m {\left\langle {\boldsymbol{\varphi}} \right|{\boldsymbol{M}}_m^\dagger {{\boldsymbol{M}}_m}\left| {\boldsymbol{\varphi}} \right\rangle = 1} }. $
量子测量同样可以采用密度算子形式来描述. 假设系统在测量前的状态是${\boldsymbol{\rho}} $,测量算子由$\left\{ {{{\boldsymbol{M}}_m}} \right\}$描述,其中i表示可能出现的测量结果,则测量得到结果i的概率为
$ p(i) = {\mathrm{tr}}({\boldsymbol{M}}_i^\dagger {{\boldsymbol{M}}_i}{\boldsymbol{\rho}} ) \text{,} $
其中tr为求迹算符,测量后的量子系统状态坍缩为
$ \frac{{{{\boldsymbol{M}}_i}{\boldsymbol{\rho}} {\boldsymbol{M}}_i^\dagger }}{{{\mathrm{tr}}({\boldsymbol{M}}_i^\dagger {{\boldsymbol{M}}_i}{\boldsymbol{\rho}} )}} . $
本节详细介绍基于QGAN的生成式信息系统的架构ECG-QGAN. ECG-QGAN能有效生成各类ECG数据,且具有高质量、多样性等特点. ECG-QGAN的总体框架如图1所示,包括数据预处理、生成器和判别器3部分. 生成器使用QBiGRU来完成生成任务,而判别器使用QCNN来完成分类任务. BiGRU是循环神经网络(recurrent neural network,RNN)的一种变体,能选择性地保留历史信息和当前信息,这在处理时间序列数据时具有优势. CNN以其强大的特征提取能力,广泛地应用于各种分类任务中. QBiGRU和QCNN是它们对应的量子版本,QBiGRU和QCNN可以通过量子叠加态和并行计算显著加快计算速度,通过量子态纠缠和量子映射,可以将经典数据映射到量子态空间进行处理,发现深层特征. 因此,QBiGRU和QCNN可以解决BiGRU和CNN在计算效率、计算速度和特征提取能力等方面所面临的挑战,而由QBiGRU和QCNN构成的QGAN还为ECG的生成提供了新的技术路线和解决方案.
ECG生成式信息系统的目标是通过对抗过程训练量子生成器G生成高质量的假ECG,以及训练量子判别器D准确判断数据的真伪. 具体来说,给定一个ECG训练集${\cal X} = {\{ {{\boldsymbol{x}}_n}\} ^N_{n=0}}$,量子生成器G由量子神经网络构成,目的是学习映射函数$G({\boldsymbol{z}})$,即将输入的1维向量噪声${\boldsymbol{z}}$转换成数据空间中的点,使其尽量模仿真实ECG数据x的数据分布. $ \boldsymbol{z}\sim P_{\mathrm{noise}}(\boldsymbol{z}) $为输入的噪声z满足的噪声分布,$\zeta $为量子生成器的变分量子电路所需要优化的参数集合. ${{\omega}} $为量子判别器的变分量子电路待优化的参数集合,但在量子生成器的参数优化过程中是个定值. 这个问题可以表达为求解式(17)
$ \mathop {\min }\limits_{\zeta } V(G) = {E_{{\boldsymbol{z}} \sim {P_{{\mathrm{noise}}}}({\boldsymbol{z}})}}\left[ {\log (1 - {D_{\boldsymbol{\omega}} }({G_{\zeta } }({\boldsymbol{z}})))} \right]. $
量子判别器D同样由量子神经网络构成,用来区分真实ECG数据${\boldsymbol{x}} $和生成的数据$G({\boldsymbol{z}})$. ${\boldsymbol{x}} \sim {P_{{\mathrm{data}}}}({\boldsymbol{x}})$为真实ECG的数据分布,在量子判别器的变分量子电路的参数优化过程中,$\zeta $作为量子生成器中待优化的参数,需要保持不变. 这个问题可以表达为求解式(18).
$\begin{split} \mathop {\max }\limits_\omega V(D) =\;& {E_{{\boldsymbol{x}} \sim {P_{{\mathrm{data}}}}({\boldsymbol{x}})}}\left[ {\log {D_{\boldsymbol{\omega}} }({\boldsymbol{x}})} \right] + \\&{E_{{\boldsymbol{z}} \sim {P_{{\text{noise}}}}({\boldsymbol{z}})}}\left[ {\log (1 - {D_{\boldsymbol{\omega}} }({G_{\zeta } }({\boldsymbol{z}})))} \right]. \end{split}$
由于生成器和判别器都是由量子神经网络构成,变分量子电路的参数的梯度下降需要利用参数位移规则找到最优解. 参数位移规则可以表达为
$\begin{split} \frac{{\partial f(\alpha )}}{{\partial {\alpha _i}}} =\;& \frac{1}{2}\left(f\left({\alpha _0},…,{\partial _i} + \frac{{\text{π}}}{2},…,{\alpha _{n - 1}}\right) -\right.\\&\left.f\left({\alpha _0},…,{\alpha _i} - \frac{{\text{π}}}{2},…,{\alpha _{n - 1}}\right)\right). \end{split}$
量子生成器和量子判别器的量子神经网络设计、实现细节和参数位移规则将在后面的章节进行详细介绍.
受量子资源限制影响,真实ECG数据需要经过预处理后方可用于量子神经网络[29],如图2所示.
数据预处理分为3个步骤:去除噪声、信号分段、数据标准化.
1)去除噪声. ECG信号的采集及数模转换过程中产生的噪声,包括IF干扰、EMG干扰及基线漂移,均会影响模型训练. 为了解决此问题,采用小波变换方法对原始心电信号进行去噪. 小波变换是一种多分辨率的时频分析方法,能够展现信号在2个域(时间和频率)上的局部特性. 它的分析窗口大小保持不变,但窗口的形状、时间分辨率和频率分辨率都可以调整. 这种方法在低频和高频部分均能实现较低的时间分辨率与较高的频率分辨率.
2)信号分段. 含有几个心跳周期的ECG数据不适合在当前NISQ设备上进行训练. 为解决这个问题,我们通过QRS检测算法定位R波峰,并利用滑动窗口技术,将长时间的ECG信号分割成单个心跳周期片段.
3)数据标准化. 分段后的ECG数据,每个数据点必须归一化为[0,π/2),以便于将经典心跳信号加载到量子态.
BiGRU是RNN的一种变体,它通过门控机制有效地解决了RNN缺乏学习长期依赖关系的能力. BiGRU通过引入门控单元(更新门和重置门),有效地控制信息流动,防止梯度消失和爆炸问题,使其能记住并利用更长时间的历史信息. QBiGRU结合了量子计算与BiGRU的优势,其量子态的叠加与纠缠特性可以帮助BiGRU更好地处理长时间跨度的依赖关系,从而提升模型在时间序列数据中的表现. 此外,相比于LSTM, GRU的门控单元更少,在当前NISQ设备上更适合与量子变分电路相结合.
生成器的结构框架如图3(a)所示,生成器的输入是满足正态分布的1维噪声向量${\boldsymbol{z}} $,将${\boldsymbol{z}} $拆分为t个一维片断向量$ {{\boldsymbol{z}}_1} $,${\boldsymbol{z}}_2 $,…,${\boldsymbol{z}}_t$,并输入到QBiGRU. QBiGRU由2个方向的QGRU单元组成,一个按照时间序列正向处理数据,另一个按照时间序列的逆向处理数据. 在正向的QGRU中,输入值${{\boldsymbol{x}}_0}$和初始化的隐藏状态${{\boldsymbol{h}}_0}$都是1维向量,并作为第1个QGRU单元的输入,其输出作为输出值1维向量${{\boldsymbol{y}}_0}$的一部分,并且作为当前QGRU单元的隐藏状态${{\boldsymbol{h}}_1}$与${{\boldsymbol{x}}_1}$一起输入到下一个QGRU单元. 在逆向的QGRU中,输入值${{\boldsymbol{x}}_t}$和初始化的隐藏状态${\boldsymbol{h}}_0'$也都是1维向量,并作为第1个QGRU单元的输入,其输出作为输出值的一部分,并且作为当前QGRU单元的隐藏状态${\boldsymbol{h}}_{1}'$与${{\boldsymbol{x}}_{t - 1}}$一起输入到下一个QGRU单元. 最后整理输出值$\left( {{\boldsymbol{y}}_0},{{\boldsymbol{y}}_1},…, {{\boldsymbol{y}}_t} \right)$,得到生成器的输出1维向量,即生成的ECG数据.
QGRU是经典GRU的量子版本. 与经典的GRU相比,QGRU用VQC替换了各门控单元中的经典神经网络. 用 VQC 替代传统神经网络可以使QGRU 的学习速度更快、收敛更稳定. 但鉴于隐藏状态${\boldsymbol{h}}$和输入${\boldsymbol{x}}$的维度不同,而VQC 的输出的维度应该与隐藏状态${\boldsymbol{h}}$的维度相匹配. 因此,需要测量的量子比特数应该是与隐藏状态${\boldsymbol{h}}$的维度相同,则用于编码输入${\boldsymbol{x}}$的量子比特的量子信息在学习过程中会被丢失和浪费. 冗余特征会消耗更多的量子比特使用量,并可能严重影响 QGRU 的学习能力,从而限制其应用潜力. 线性层可以解决这一问题,它可以将输入特征映射到目标维度,以获得目标维度中的特征表示. QGRU的结构如图3(b)所示,相应的前向传递方程为:
$ {{{z}}_t} = \sigma ({L_{{\mathrm{out}}}}(VQ{C_1}({L_{{\mathrm{in}}}}([{{\boldsymbol{h}}_{t - 1}},{{\boldsymbol{x}}_t}]))))\text{,} $
$ {{{r}}_t} = \sigma ({L_{{\mathrm{out}}}}(VQ{C_2}({L_{{\mathrm{in}}}}([{{\boldsymbol{h}}_{t - 1}},{{\boldsymbol{x}}_t}]))))\text{,} $
$ {\widetilde{\boldsymbol{h}}_{t - 1}} = {\mathrm{tanh}}({L_{{\mathrm{out}}}}(VQ{C_3}({L_{{\mathrm{in}}}}([{{\boldsymbol{r}}_t} \cdot {{\boldsymbol{h}}_{t - 1}},{{\boldsymbol{x}}_t}]))))\text{,} $
$ {{\boldsymbol{h}}_t} = (1 - {{{z}}_t}) \cdot {{\boldsymbol{h}}_{t - 1}} + {{{z}}_t} \cdot {\widetilde{\boldsymbol{h}}_{t - 1}}. $
其中,下标t表示时间步长;${{\boldsymbol{x}}_t}$和${{\boldsymbol{h}}_{t - 1}}$分别代表输入数据和隐藏状态;[ ]表示向量串联; ${{{z}}_t}$,${{{r}}_t}$分别表示QGRU中的更新门、重置门;${{\boldsymbol{h}}_t}$表示新的隐藏状态;$ {{\widetilde{\boldsymbol{h}}_{t - 1}}}$表示新的候选状态;${L_{{\mathrm{in}}}}$和${L_{{\mathrm{out}}}}$表示线性层;$VQ{C_{1,2,3}}$表示变分量子电路;$\sigma \left( {} \right)$和${\mathrm{tanh}}\left( {} \right)$分别表示Sigmoid和tanh激活函数;“$ \cdot $”表示 Hadamard 乘积(逐元素乘积);“$ + $”表示逐元素相加.
近期量子设备的量子比特数量有限. 因此,设计需要更少量子比特的变分量子电路至关重要. 在变分量子电路的设计中,我们采用了一种张量网络量子电路,称为“矩阵乘积状态(MPS)量子电路”. 量子生成器可使用具有较少量子比特的MPS量子电路来构建QGRU中的VQC. 图4展示了MPS量子电路的结构,它类似于一棵具有 N 个节点的最大不平衡树,每个节点都是一个量子模拟.
具体地说,MPS量子电路由3部分组成,分别为量子编码层、变分层和测量. 量子编码层采用振幅编码将经典数据编码至量子态. 振幅编码可以高效地表示大量的经典信息,对于一个有N个元素的经典数据,只需要$ \log N $个量子比特,其表示如式(24)所示.
$ \left| {\boldsymbol{q}} \right\rangle = \sum\limits_{i = 0}^{{2^n} - 1} {{q_i}\left| i \right\rangle } \text{,} $
其中${q_i}$是向量q的第i个元素,$\left| i \right\rangle $是i的二进制表示对应的量子态. MPS量子电路的量子变分层是通过Ansatz块实现的(Ansatz指量子电路结构). 每个Ansatz块包含2个量子位,相当于1个节点. 每个节点的最上面的量子位在完成旋转和纠缠后进行测量,下面的量子位流到下一个节点. 量子变分层的旋转和量子比特之间的纠缠由${{\boldsymbol{R}}_x}({{\theta}} )$旋转门、${{\boldsymbol{R}}_y}({{\theta}} )$旋转门和CNOT门实现,即式(7)~(9)所示. 经过量子变分层后,整个电路的量子态为$\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle = {\boldsymbol{U}}({{\theta }})\left| {\boldsymbol{q}} \right\rangle $,$\left| {{\boldsymbol{U}}(\theta )} \right\rangle $表示变分层所有的操作,包括旋转门和CNOT门.
量子测量是测量量子态$\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle $从而得到经典数据. 按照MPS量子电路特点,对其中一个量子位s进行测量得到$ p(m) = \left\langle {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right|{P_m} \otimes {\boldsymbol I}\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle $,其他量子位流动到下一个节点,即剩下量子态,表示如式(25)所示.
$\begin{split} {\boldsymbol{\rho}} =\;& \frac{{{\mathrm{tr}}_{s}({P_m} \otimes {\boldsymbol I}\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle \left\langle {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right|{P_m} \otimes {\boldsymbol I})}}{{{\mathrm{tr}}({P_m} \otimes {\boldsymbol I}\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle \left\langle {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right|{P_m} \otimes {\boldsymbol I})}} = \\&\frac{{{\mathrm{tr}}_{s}({P_m} \otimes {\boldsymbol I}\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle \left\langle {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right|)}}{{\left\langle {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right|{P_m} \otimes {\boldsymbol I}\left| {{\boldsymbol{\varphi}} ({\boldsymbol{q}})} \right\rangle }}.\end{split} $
重复测量操作,直至所有量子位都测量完成,得到MPS量子电路的输出结果.
判别器是用来区分真实样本和生成样本的,其结构如图5所示. QCNN是CNN的量子版本,其优势是利用量子态的并行处理能力,可以在较少的计算资源下,高效地提取和处理数据的特征. 并且量子纠缠态和量子叠加态的特性使量子CNN在处理数据时,能够更好地捕捉数据中的复杂特征,提高分类的准确性. QCNN采用长度为n的量子滤波器,将滤波器内的ECG数据1维向量$ ({{\boldsymbol{x}}_j},{{\boldsymbol{x}}_{j + 1}},…,{{\boldsymbol{x}}_{j + n}}) $映射到$N = \log n$量子位的量子态$\left| {\boldsymbol{\psi}} \right\rangle = \displaystyle\sum\limits_{k=j}^{j + n} {{{\boldsymbol{x}}_k}\left| i \right\rangle } $. 经过VQC实现量子卷积操作,然后对其测量,测量结果作为特征的一部分. 最后经过全连接层,判断输入的数据是真实样本还是生成样本.
QCNN采用TTN量子电路,其结构如图6所示. 它拥有与二叉树相同的结构,每个节点都用1个Ansatz块进行模拟. 在TTN量子电路中,每2个量子位构建1个Ansatz块. 每1个Ansatz块会对2个量子位施加${{\boldsymbol{R}}_x}({{\theta}} )$旋转门和CNOT门,实现量子位旋转和纠缠. 受控的量子位保留到下一个Ansatz块中,控制的量子位被舍弃,直至只剩下一个量子位,对其进行测量,测量结果作为输出. TTN量子电路能够用较少的参数有效地表示那些具有层次结构或局部关联的量子态,从而减少对存储和计算资源的需求.
为了防止MPS量子电路和TTN量子电路出现过拟合现象,我们对其引入了量子Dropout策略. 图7是经典Dropout和量子Dropout的图示. 经典Dropout用于防止经典深度学习模型中的过拟合,它包括在训练阶段的每次迭代中随机从网络中删除神经元及其所有传入和传出连接,如图7(a)虚线所示. 量子Dropout则是将单个量子位旋转门与以该单旋转量子位为控制位的所有后续纠缠门一起随机丢弃,如图7(b)虚线框所示.
GAN的损失函数如式(26)所示. 生成器 G 接收随机噪声$ {\boldsymbol{z}} \sim {P_{{\mathrm{noise}}}}({\boldsymbol{z}}) $,并尝试生成与训练数据分布相似的合成数据. 鉴别器 D 尝试确定输入的数据是真实的还是生成的. 生成器的目标是最大化鉴别器的错误率,而鉴别器的目标是最小化错误率. ${\boldsymbol{x}} \sim {P_{{\mathrm{data}}}}({\boldsymbol{x}})$来自真实数据.
$ \begin{split}\mathop {\min }\limits_G \mathop {\max }\limits_D V(G,D) =\;& {E_{{\boldsymbol{x}} \sim {P_{{\mathrm{data}}}}({\boldsymbol{x}})}}[\log D({\boldsymbol{x}})]+ \\ &{E_{{\boldsymbol{z}} \sim {P_{{\mathrm{noise}}}}({\boldsymbol{z}})}}[\log (1 - D(G({\boldsymbol{z}})))]. \end{split} $
最初的 GAN 工作已经证明了 GAN 在训练过程中的全局最优性和收敛性. 然而,它仍然凸显了训练 GAN 时可能出现的不稳定问题,如梯度消失. 梯度消失是由直接优化式(26)中的损失引起的. 当 D 达到最优时,G 的优化方程可以转换为最小化真实数据分布$({P_{{\mathrm{data}}}})$和生成器分布$({P_{{\mathrm{noise}}}})$之间的 Jensen-Shannon (JS) 散度为
$ {L_G} = 2JS({P_{{\mathrm{data}}}}\parallel {P_{{\mathrm{noise}}}}) - 2\log 2. $
当${P_{{\mathrm{data}}}}$和${P_{{\mathrm{noise}}}}$之间没有重叠时,${L_G}$保持恒定$\log 2$,表明在这种情况下使用此损失的 G 的梯度接近 0. 仅当${P_{{\mathrm{data}}}}$和${P_{{\mathrm{noise}}}}$有大量重叠时,G 的非零梯度才存在. 实际上,${P_{{\mathrm{data}}}}$和${P_{{\mathrm{noise}}}}$不相交或重叠可忽略不计的可能性相当高. 为了解决这个问题,人们一直在努力重新制定具有完全不同目标的 GAN 训练,以获得更好的理论保证. 最成功的框架之一是 WGAN [30]. WGAN 最小化 Wasserstein 距离:
$ W({P_{{\mathrm{data}}}},{P_{{\mathrm{noise}}}}) = {\inf _{\gamma \sim \varPi ({P_{{\mathrm{data}}}},{P_{{{\mathrm{noise}}}}})}}{E_{({\boldsymbol{x}},{\boldsymbol{y}}) \sim \gamma }}[\parallel {\boldsymbol{x}} - {\boldsymbol{y}}\parallel ], $
其中$\varPi ({P_{{\mathrm{data}}}},{P_{{\mathrm{noise}}}})$是联合分布,并从中采样$({\boldsymbol{x}},{\boldsymbol{y}})$,x是真实样本,${\boldsymbol{y}} = G({\boldsymbol{z}})$是生成样本,并计算这2种样本之间的距离$\parallel {\boldsymbol{x}} - {\boldsymbol{y}}\parallel $. 在所有可能的联合分布中,存在一个下界,能够对这个期望值取到的下界为$ {\inf\limits_{\gamma \sim \varPi ({P_{{\mathrm{data}}}},{P_{{{\mathrm{noise}}}}})}}{E_{({\boldsymbol{x}},{\boldsymbol{y}}) \sim \gamma }}[\parallel {\boldsymbol{x}} - {\boldsymbol{y}}\parallel ] $. 相较于KL散度和JS散度,Wasserstein距离的优势在于它能有效衡量即使2个分布没有重叠情况下的距离远近. 其对应损失函数如式(29)(30)所示.
$ {L_G} = - {E_{{\boldsymbol{z}} \sim {P_{{\mathrm{noise}}}}}}[{f_{\mathrm{w}}}(G({\boldsymbol{z}}))]\text{,} $
$ {L_D} = {E_{{\boldsymbol{z}} \sim {P_{{{\mathrm{noise}}}}}}}[{f_{\mathrm{w}}}(G({\boldsymbol{z}}))] - {E_{{\boldsymbol{x}} \sim {P_{{\mathrm{data}}}}}}[{f_{\mathrm{w}}}({\boldsymbol{x}})]. $
接下来我们将采用参数移位规则(parameter shift rule)实现变分量子电路中的梯度下降. 参数移位规则指出,我们可以通过简单地将参数移位2次并计算2个输出之间的差来计算某些量子电路中每个参数的梯度,而无需改变电路的结构或使用任何辅助量子位. 假设一个m量子位的量子电路由n个参数${\theta _0},{\theta _1},…,{\theta _{n - 1}}$进行参数化,$\theta =\{{\theta _0},{\theta _1},…,{\theta _{n - 1}}\} $,该电路的测量期望值可以用电路函数表示为
$ f({{\theta}} ) = \left\langle {\boldsymbol{\psi}} \right|{\boldsymbol{U}}{({\theta _i})^\dagger }\widehat {\boldsymbol{OU}}({\theta _i})\left| {\boldsymbol{\psi}}\right\rangle , $
其中${\theta _i}$是要计算梯度的标量参数,${\boldsymbol{U}}({\theta _i})$是${\theta _i}$所在的门,${\boldsymbol{U}}({\theta _i})$之前的幺正矩阵吸收到$ \left\langle {\boldsymbol{\psi}} \right| \cdot \left| {\boldsymbol{\psi}} \right\rangle $中,${\boldsymbol{U}}({\theta _i})$之后的幺正矩阵用$\widehat {\boldsymbol{O}}$表示. VQC中使用的量子门可以写成${\boldsymbol{U}}({\theta _i}) = {{\mathrm{e}}^{ - \frac{{\mathrm{i}}}{2}{\theta _i}{\boldsymbol{H}}}}$的形式. 这里H是U的埃尔米特生成器,只有2个唯一的特征值+1和–1(H的特征值可以是±r,但为了简单起见,我们假设它是±1). 这样,电路函数f相对于${\theta _i}$的梯度为
$ \begin{split}\frac{{\partial f({{\theta}} )}}{{\partial {\theta _i}}} =\;& \frac{1}{2}\left(f\left({\theta _0},…,{\theta _i} + \frac{{\text{π}}}{2},…,{\theta _{n - 1}}\right)- \right.\\ &\left.f\left({\theta _0},…,{\theta _i} - \frac{{\text{π}}}{2},…,{\theta _{n - 1}}\right)\right). \\ \end{split} $
接下来将${\boldsymbol{U}}({\theta _i})$具体到${{\boldsymbol{R}}_{\boldsymbol{x}}}({\theta _i})$. 假设${\boldsymbol{U}}({\theta _i}) = {{\boldsymbol{R}}_{\boldsymbol{x}}}({\theta _i})$,${{\boldsymbol{R}}_x}(\alpha ) = {{\mathrm{e}}^{ - \frac{{\mathrm{i}}}{2}\alpha {\boldsymbol{X}}}}$,其中X是Pauli-X矩阵. 首先,
$ \begin{split}{{\boldsymbol{R}}_x}(\alpha ) =\;& {{{\mathrm{e}}}^{ - \tfrac{{\mathrm{i}}}{2}\alpha {\boldsymbol{X}}}} = {\sum\limits_{j = 0}^\infty {\left( - \frac{{{\mathrm{i}}\alpha }}{2}\right)} ^j}{{\boldsymbol{X}}^j}/k! =\\ &{\sum\limits_{j = 0}^\infty {\left( - \frac{{{\mathrm{i}}\alpha }}{2}\right)} ^{2j}}{{\boldsymbol{X}}^{2j}}/(2k)! +\\ &{\sum\limits_{j = 0}^\infty {\left( - \frac{{{\mathrm{i}}\alpha }}{2}\right)} ^{2j + 1}}{{\boldsymbol{X}}^{2j + 1}}/(2k + 1)! = \\ &{\sum\limits_{j = 0}^\infty {{{( - 1)}^j}\left(\frac{\alpha }{2}\right)} ^{2j}}{\boldsymbol{I}}/(2k)!- \\ &{\mathrm{i}}{\sum\limits_{j = 0}^\infty {{{( - 1)}^j}\left(\frac{\alpha }{2}\right)} ^{2j + 1}}{\boldsymbol{X}}/(2k + 1)! =\\ &\cos (\alpha /2){\boldsymbol{I}} - {\mathrm{i}}\sin (\alpha /2){\boldsymbol{X}}. \\ \end{split} $
当$\alpha = \dfrac{{\text{π}}}{2}$时,${{\boldsymbol{R}}_x}\left( \pm \dfrac{{\text{π}} }{2}\right) = \dfrac{1}{{\sqrt 2 }}({\boldsymbol{I}} \mp {\mathrm{i}}{\boldsymbol{X}})$,由式(32),$ {{\boldsymbol{R}}}_{x}(\alpha ){{\boldsymbol{R}}}_{x}(\beta )={{\boldsymbol{R}}}_{x}(\alpha +\beta) $和$\dfrac{{\partial {{\boldsymbol{R}}_x}(\alpha )}}{\alpha } = - \dfrac{{\mathrm{i}}}{2}{\boldsymbol{X}}{{\boldsymbol{R}}_x}(\alpha )$可得:
$ \begin{split}\frac{{\partial f({{\theta}} )}}{{\partial {\theta _i}}} =\;& \left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }{\left( - \frac{{\mathrm{i}}}{2}{\boldsymbol{X}}\right)^\dagger }\widehat {\boldsymbol{O}}{{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle +\\ &\left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }\widehat {\boldsymbol{O}}\left( - \frac{{\mathrm{i}}}{2}{\boldsymbol{X}}\right){{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle= \\ &\frac{1}{4}(\left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }{({\boldsymbol{I}} - {\mathrm{i}}{\boldsymbol{X}})^\dagger }\widehat {\boldsymbol{O}}({\boldsymbol{I}} - {\mathrm{i}}{\boldsymbol{X}}){{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle -\\ &\left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }{({\boldsymbol{I}} + {\mathrm{i}}{\boldsymbol{X}})^\dagger }\widehat {\boldsymbol{O}}({\boldsymbol{I}} + {\mathrm{i}}{\boldsymbol{X}}){{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle ) =\\ &\frac{1}{2}(\left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }{{\boldsymbol{R}}_x}{\left(\frac{{\text{π}}}{2}\right)^\dagger }\widehat {\boldsymbol{O}}{{\boldsymbol{R}}_x}\left(\frac{{\text{π}} }{2}\right){{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle -\\ &\left\langle {\boldsymbol{\psi}} \right|{{\boldsymbol{R}}_x}{({\theta _i})^\dagger }{{\boldsymbol{R}}_x}{\left( - \frac{{\text{π}} }{2}\right)^\dagger }\widehat {\boldsymbol{O}}{{\boldsymbol{R}}_x}\left( - \frac{{\text{π}}}{2}\right){{\boldsymbol{R}}_x}({\theta _i})\left| {\boldsymbol{\psi}} \right\rangle )= \\ &\frac{1}{2}\left(f\left({\theta _0},…,{\theta _i} + \frac{{\text{π}} }{2},…,{\theta _{n - 1}}\right) - \right.\\&\left.f\left({\theta _0},…,{\theta _i} - \frac{{\text{π}}}{2},…,{\theta _{n - 1}}\right)\right).\\[-1pt]\end{split} $
不失一般性,式(34)同时适用于${{\boldsymbol{R}}_x}$,${{\boldsymbol{R}}_y}$,${{\boldsymbol{R}}_z}$. 当然,在某些情况下,一个参数存在于多个门中. 在这种情况下,我们只需要分别计算这些门中参数的梯度,并将梯度相加即可得到该参数的梯度.
我们选用4个公共医疗心跳数据集进行实验来评估我们提出的ECG-QGAN,我们将这4个数据集进行裁剪,然后保留R峰的前0.4 s和后0.5 s的心电信号为单个样本,单个样本总持续时间为0.9 s,然后再将其重采样至256维心电信号. 我们从这4个数据库中随机抽取了500个标记为N,S,V,F,Q的心跳样本进行训练,总共2 500个训练样本. 数据集的详细介绍如表1所示.
1)MIT-BIH Arrhythmia Database[31]. 双通道动态ECG记录,由BIH Arrhythmia Laboratory在1975—1979年间研究的47名受试者中获得.
2)MIT-BIH ST Change Database[32]. 28个不同长度的ECG记录,其中大多数是在运动压力测试期间记录的,表现出短暂的ST段压低.
3)European ST-T Database[33]. 79名受试者动态ECG记录的注释摘录,旨在评估ST段和T波变化分析算法.
4)Sudden Cardiac Death Holter Database[34]. 一个由hysioNet建立的心脏性猝死数据库.
为了验证本文提出的ECG-QGAN的性能,我们将其部署在Pytorch和PennyLane上进行模拟实验. 本文研究的所有实验均在Pytorch 1.12.0、PennyLane 0.20.1、Python 3.9.13中进行. 环境为:操作系统为Windows 10,CPU为Intel®Core™i7-8700 CPU 3.10 GHz,GPU为NVIDIA GeForce GTX 1050ti,内存为16 GB,存储为1 TB.
GAN可能很难评估,研究人员尚未就哪些指标最能反映 GAN 性能达成一致. 文献[35]提出了很多指标,其中大多数适用于计算机视觉领域. 适当评估时间序列 GAN 的工作仍在进行中. 我们可以将评估指标分为2类:定性和定量. 定性评估是通过检查 GAN 生成的样本进行人类视觉评估的另一个术语. 然而,由于缺乏合适的客观评估指标,这不能被视为对GAN性能的全面评估. 定量评估包括使用与用于时间序列分析的统计测量相关的指标和相似性测量,例如皮尔逊相关系数 (PCC)、百分比均方根差 (PRD)和均方根误差 (RMSE). 这些指标是最常用于时间序列评估的度量,因此可用作我们提出的模型性能指标,因为它们可以反映训练数据和合成生成数据之间的稳定性. 我们在式(35)~(37)中展示了这些常见公式.
$ PCC = \dfrac{{\displaystyle\sum\limits_{i = 1}^N {({x_i} - \widetilde x)({y_i} - \widetilde y)} }}{{\sqrt {\displaystyle\sum\limits_{i = 1}^N {{{({x_i} - \widetilde x)}^2}} \displaystyle\sum\limits_{i = 1}^N {{{({y_i} - \widetilde y)}^2}} } }} \text{,} $
$ PRD = \sqrt {\dfrac{{\displaystyle\sum\limits_{i = 1}^N {{{({x_i} - {y_i})}^2}} }}{{\displaystyle\sum\limits_{i = 1}^N {{{({x_i})}^2}} }}} \text{,} $
$ RMS E = \sqrt {\frac{1}{N}\sum\limits_{i = 1}^N {{{({x_i} - {y_i})}^2}} }. $
在式(35)~(37)中,${x_i}$是时间序列x在时间i的实际值,${y_i}$是时间序列y在时间i的生成值. $\widetilde x$和$\widetilde y$分别表示x和y的平均值.
本文根据3.3节和3.4节提出的基于MPS量子电路和基于TTN量子电路,实现量子比特之间的纠缠,并应用于ECG-QGAN,旨在利用量子纠缠态来增强生成能力. 由于本文将QBiGRU和QCNN一起应用于GAN中,所以本文重点与BiGRU-CNN,BiLSTM-CNN进行比较. 其他的经典模型和量子模型通过4.2节提到的评价指标进行比较,包括DCGAN[36]、量子补丁GAN(patch GAN)[26]和QCGAN-ECG[28]. 为了更准确地评估ECG-QGAN的性能,本文进行一系列的实验和分析.
为了验证Wasserstein距离对于JS散度和KL散度的优势,本文首先进行了一个简单的实验. 分别采用Wasserstein距离、JS散度和KL散度进行少量ECG数据的生成任务,并通过生成器的损失值(Loss)变化曲线,比较Wasserstein距离、JS散度和KL散度的优劣. 在图8中,采用Wasserstein距离的生成器的Loss虽然在训练开始阶段会出现波动,但随着训练轮次增加,逐渐平稳且收敛. 相较之下采用JS散度和KL散度的生成器的Loss曲线可能在某些时间点出现突变或不稳定,从训练开始到训练结束一直出现剧烈波动,且没有收敛,这正验证了3.5节所述的训练经典GAN会出现梯度消失等不稳定问题,即当2个分布的支撑集没有重叠或者重叠非常少时,KL散度可能会变得没有意义,而JS散度虽然能给出一定的度量值,但它会退化为一个常数,无法准确反映2个分布的远近. 而采用Wasserstein距离的WGAN可以避免梯度消失问题的出现. 因为它衡量的是将数据从一个分布“移动”到另一个分布所需要的最小平均距离,提供了一种更平滑和更稳定的度量方式. 综上所述,Wasserstein距离在处理时间序列数据时,相较于KL散度和JS散度,展现出了更高的稳定性和更平滑的变化趋势,从而可以在训练过程中提供更稳定的梯度.
接着本文将准备5类心电信号(N,S,V,F,Q)各500个进行不同种类心跳生成的模拟实验,每个样本的维度为256. 我们将迭代总数设置为 Epoch = 1500,判别器和生成器的学习率为 ${\mu _D}={\mu _G} = 0.000\;1$,batch_size = 64.在生成器的量子电路中,我们使用的量子比特数为4,每个方向的QGRU单元数量为4. 在判别器的量子电路中,使用的量子比特数也为4. BiGRU-CNN GAN和BiLSTM-CNN GAN的模型架构尽量与ECG-QGAN相同,但参数量远远高于ECG-QGAN. 在整个训练过程中,在生成N,V,F,Q类ECG时,判别器的Loss如图9所示,生成S类ECG时,判别器的损失值如图10所示.
为了验证ECG-QGAN模型生成器和判别器之间的对抗博弈,我们在图9和图10中展示了判别器Loss的变化曲线. 我们采用Wasserstein 距离代替初始GAN的JS散度,因此可以通过判别器Loss的变化验证判别器和生成器之间的对抗性. 当在图9和图10中仅关注ECG-QGAN时,可以看到,ECG-QGAN在Epoch处于0~100时,判别器的Loss持续下降,表明其判别能力弱于生成能力,需要增强其判别器的学习能力. 随着Epoch的增加,判别器的Loss开始缓慢降低,说明生成器和判别器的学习能力都在提高,但生成器的学习能力提高的速度低于判别器. 当Epoch=400时,判别器的Loss开始趋于平稳,说明生成器和判别器逐渐接近纳什均衡.
除了上述2种经典模型,我们还选用DCGAN,QPatch-GAN,QCGAN-ECG作为基准模型进行比较. 通过4.2节的3个评价指标来分析ECG-QGAN的性能. 我们严格设置相当的模型参数,使用相同的数据集、训练轮次以及学习率,从而确保实验结果的公平性. 表2~4展示了基于PCC,PRD,RMSE指标的对比结果. PCC数值越高表示生成的ECG信号质量越高;而PRD和RMSE数值越低,表示生成的ECG信号质量越高. 在表2中,ECG-QGAN占有4类ECG信号的最佳表现. 尽管Q类ECG信号未能达到最佳表现,但仍优于其他模型. 在表3中,ECG-QGAN也占有4类ECG信号的最佳表现. F类ECG信号未能达到最佳表现,但接近最佳. 在表4中,ECG-QGAN仅占有2类ECG信号的最佳表现. 然而,综合3个评价指标,ECG-QGAN的整体生成能力优于其他模型.
为了直观地显示生成ECG数据的质量,图11对比了3种模型生成的ECG数据与数据集中的真实ECG数据. 第1列为真实ECG数据,第2列至第4列分别是ECG-QGAN、BiGRU-CNN GAN和BiLSTM-CNN GAN生成的数据. 每行展示不同种类的ECG数据. 从图11可以看出,与真实的ECG数据相比,ECG-QGAN生成的ECG信号都保留了不同类别ECG的特征,不管是正常的心跳还是异常的心跳. 这表明本文提出的模型无需医疗专家手动标注,其本身就已经带有不同类别的标注.此外,本文提出的模型是基于真实ECG数据生成,其本身不含有任何患者的隐私. 因此,对于生成的ECG数据在使用上没有限制. 最后,与其他2种经典生成对抗网络相比,本文提出的模型生成的ECG信号质量更高. 这一成果归功于变分量子电路,量子门操作实现量子间的计算和纠缠从而提高了模型的学习能力和特征提取能力.
选取QCGAN-ECG,BiGRU-CNN GAN,BiLSTM-CNN GAN作为基准模型,以分类性能来评估4种模型的生成质量. 选取Inception V3作为分类模型,并从预处理后的MIT-BIH Arrhythmia Database,MIT-BIH ST Change Database,European ST-T Database,Sudden Cardiac Death Holter Database数据集中抽取2 500个样本,其中每类ECG信号各500个样本. 我们进行50轮训练,选择SGD优化器,学习率设置为0.01,损失函数选择交叉熵损失函数,训练50轮后,整个分类准确率达到90%,保存模型以及参数. 我们将3种基准模型和ECG-QGAN的生成结果放入预训练的InceptionV3进行分类,其分类结果如图12所示. ECG-QGAN的准确率达到0.826,优于基准模型. 同时,所有类别的ECG信号准确率都接近于0.8,表明ECG-QGAN能有效生成5种类别ECG数据. 这进一步验证了ECG-QGAN所生成的ECG数据可以用于训练ECG分类算法.
量子计算机的量子计算过程中不可避免地存在量子噪声. 量子噪声带来的扰动会导致信息熵随着量子态从纯态到混合态的变化而增加. 为了验证模型的鲁棒性,我们分析了4种量子噪声:比特翻转噪声、相位翻转噪声、幅度阻尼噪声和去极化噪声引起的扰动的影响. 它们的Kraus算子分别如式(38)~(41)所示.
$ {{\boldsymbol{E}}_{0}} = \sqrt {1 - \gamma } {\boldsymbol{I}},{{\boldsymbol{E}}_{1}} = \sqrt \gamma {{\boldsymbol{\sigma}} _x}\text{,} $
$ {{\boldsymbol{E}}_0} = \sqrt {1 - \gamma } {\boldsymbol{I}},{{\boldsymbol{E}}_1} = \sqrt \gamma {{\boldsymbol{\sigma}} _z}\text{,} $
$ {{\boldsymbol{E}}_{0}} = \left( {\begin{array}{*{20}{c}}1&0 \\ 0&{\sqrt {1 - \gamma } } \end{array}} \right),{{\boldsymbol{E}}_{1}} = \left( {\begin{array}{*{20}{c}}0&{\sqrt \gamma } \\ 0&0 \end{array}} \right)\text{,} $
$ \begin{split}&{{\boldsymbol{E}}_0} = \sqrt {1 - \gamma } {\boldsymbol{I}},{{\boldsymbol{E}}_1} = \sqrt {{\gamma /3}} {{\boldsymbol{\sigma}} _x} ,\\&{{\boldsymbol{E}}_2} = \sqrt {{\gamma /3}} {{\boldsymbol{\sigma}} _y},{{\boldsymbol{E}}_{3}} = \sqrt {{\gamma /3}} {{\boldsymbol{\sigma}} _z}\end{split} . $
在式(38)~(41)中,$\gamma $分别表示比特翻转噪声、相位翻转噪声、幅度阻尼噪声和去极化噪声的0~1范围内的噪声参数. I 是单位矩阵,${{\boldsymbol{\sigma}} _x}$,${{\boldsymbol{\sigma}} _y}$,${{\boldsymbol{\sigma}} _z}$则是另外的3个泡利矩阵. 保真度用于测量2个量子态之间的距离,它被定义为式(42).
$ F\left( {{\boldsymbol{\rho }},{\boldsymbol{\sigma}} } \right) \equiv {\mathrm{tr}}\left(\sqrt {{\boldsymbol{\rho}}^{1/2}{\boldsymbol{\sigma}} {{\boldsymbol{\rho}}^{1/2}}}\right). $
将4种量子噪声分别加入到训练过程中,得到实验错误率. 分别在比特翻转、相位翻转、幅度阻尼和去极化4种噪声下的4组理论错误率与实验错误率进行比较,如图13所示. 无论理论错误率如何变化,实验错误率都低于理论错误率. 由此可见,量子电路的训练过程具有抗量子噪声的能力.
本文提出了一种运用量子技术生成ECG数据的生成式信息系统,称为ECG-QGAN. ECG作为一种时间序列数据,对时间具有依赖性,而递归神经网络非常适用于这类任务. ECG-QGAN的生成器是一种递归神经网络,同时生成器和判别器采用基于矩阵乘积状态和树形张量网络的变分量子电路. 这不仅提升了模型的学习和特征提取能力,还节省了量子资源. 实验证明,ECG-QGAN在生成各类ECG数据时均表现出令人满意的效果. 生成的ECG数据不涉及患者隐私且无需人工标注,有助于解决ECG数据量不足的问题. 与其他GAN相比,它可以更快达到纳什均衡,并在多个评价指标上获得最佳的表现. 此外,我们在分类任务中使用了ECG-QGAN的生成结果,在经过预训练的模型检验后,其分类准确率可以达到0.826. 这证明ECG-QGAN可以用于扩充ECG数据集,解决ECG分类算法缺乏足够的数据量训练的问题. 在未来的工作中,我们将专注于开发更节省量子资源且更高效的量子变分电路.
作者贡献声明:瞿治国提出研究思路和模型,负责论文写作;陈韦龙负责完成实验和论文写作;孙乐补充实验并总结实验结果,修改论文;刘文杰和张彦春负责修改论文.

参考文献 引证文献
排序方式:
1
Mensah G A, Fuster V, Murray C J L, et al. Global burden of cardiovascular diseases and risks, 1990-2022[J]. Journal of the American College of Cardiology, 2023, 82(25): 2350−2473
2
闫璟辉,宗成庆,徐金安. 中文医疗文本中的嵌套实体识别方法[J]. 软件学报,2024,35(6):2923−2935
Yan Jinghui, Zong Chengqing, Xu Jin’an. Configurable text-based image editing by autoencoder-based generative adversarial networks[J]. Journal of Software, 2024, 35(6): 2923−2935 (in Chinese)
3
蔡德润,李红燕. 基于度量学习的无监督域适应方法及其在死亡风险预测上的应用[J]. 计算机研究与发展,2022,59(3):674−682
Cai Derun, Li Hongyan. A metric learning based unsupervised domain adaptation method with its application on mortality prediction[J]. Journal of Computer Research and Development, 2022, 59(3): 674−682 (in Chinese)
4
张颖超, 茅丹, 胡凯. 压缩传感理论在心电图信号恢复问题上的研究[J]. 计算机研究与发展,2014,51(5):1018−1027
Zhang Yingchao, Mao Dan, Hu Kai. ECG signal recovery problem based on compressed sensing theory[J]. Journal of Computer Research and Development, 2014, 51(5): 1018−1027 (in Chinese)
5
Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[J]. Communications of the ACM, 2020, 63(11): 139−144
6
吴福祥,程俊. 基于自编码器生成对抗网络的可配置文本图像编辑. 软件学报[J],2022,33(9):3139−3151
Wu Fuxiang, Cheng Jun. Configurable text-based image editing by autoencoder-based generative adversarial networks[J]. Journal of Software, 2022, 33(9): 3139−3151 (in Chinese)
7
于海涛,杨小汕,徐常胜. 基于多模态输入的对抗式视频生成方法[J]. 计算机研究与发展,2020,57(7):1522−1530
Yu Haitao, Yang Xiaoshan, Xu Changsheng. Antagonistic video generation method based on multimodal input[J]. Journal of Computer Research and Development, 2020, 57(7): 1522−1530 (in Chinese)
8
蒋斌,刘虹雨,杨超. 一种基于局部属性生成对抗网络的人脸修复算法[J]. 计算机研究与发展,2019,56(11):2485−2493
Jiang Bin, Liu Hongyu, Yang Chao. A face inpainting algorithm with local attribute generative adversarial networks[J]. Journal of Computer Research and Development, 2019, 56(11): 2485−2493 (in Chinese)
9
戴臣超,王洪元,倪彤光. 基于深度卷积生成对抗网络和拓展近邻重排序的行人重识别[J]. 计算机研究与发展,2019,56(8):1632−1641
Dai Chenchao, Wang Hongyuan, Ni Tongguang. Person reidentification based on deep convolutional generative adversarial network and expanded neighbor reranking[J]. Journal of Computer Research and Development, 2019, 56(8): 1632−1641 (in Chinese)
10
Brophy E, Wang Zhengwei, She Qi, et al. Generative adversarial networks in time series: A systematic literature review[J]. ACM Computing Surveys, 2023, 55(10): 1−31
11
Boixo S, Isakov S V, Smelyanskiy V N, et al. Characterizing quantum supremacy in near-term devices[J]. Nature Physics, 2018, 14(6): 595−600
12
Biamonte J, Wittek P, Pancotti N, et al. Quantum machine learning[J]. Nature, 2017, 549(7671): 195−202
13
Preskill J. Quantum computing in the NISQ era and beyond[J]. Quantum, 2018, 2: 79−88
14
Schuld M, Sinayskiy I, Petruccione F. An introduction to quantum machine learning[J]. Contemporary Physics, 2015, 56(2): 172−185
15
Ciliberto C, Herbster M, Ialongo A D, et al. Quantum machine learning: A classical perspective[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2018, 474(2209): 20170551
16
李盼池,周红岩. 基于受控Hadamard门的量子神经网络模型及算法[J]. 计算机研究与发展,2015,52(1):211−220
Li Panchi, Zhou Hongyan. Model and algorithm of quantum neural network based on the controlled Hadamard gates[J]. Journal of Computer Research and Development, 2015, 52(1): 211−220 (in Chinese)
17
Benedetti M, Lloyd E, Sack S, et al. Parameterized quantum circuits as machine learning models[J]. Quantum Science and Technology, 2019, 4(4): 043001
18
Delaney A M, Brophy E, Ward T E. Synthesis of realistic ECG using generative adversarial networks[J]. arXiv preprint, arXiv: 1909.09150, 2019
19
Golany T, Radinsky K. PGANs: Personalized generative adversarial networks for ECG synthesis to improve patient-specific deep ECG classification[C] // Proc of the AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2019: 557−564
20
Golany T, Radinsky K, Freedman D. SimGANs: Simulator-based generative adversarial networks for ECG synthesis to improve deep ECG classification[C] // Proc of the Int Conf on Machine Learning. New York: ACM, 2020: 3597−3606
21
Vo K, Naeini E K, Naderi A, et al. P2E-WGAN: ECG waveform synthesis from PPG with conditional Wasserstein generative adversarial networks[C] //Proc of the 36th Annual ACM Symp on Applied Computing. New York: ACM, 2021: 1030−1036
22
Adib E, Fernandez A S, Afghah F, et al. Synthetic ECG signal generation using probabilistic diffusion models[J]. IEEE Access, 2023, 11: 75818−75828
23
Lloyd S, Weedbrook C. Quantum generative adversarial learning[J]. Physical Review Letters, 2018, 121(4): 040502
24
Dallaire-Demers P L, Killoran N. Quantum generative adversarial networks[J]. Physical Review A, 2018, 98(1): 012324
25
Benedetti M, Garcia-Pintos D, Perdomo O, et al. A generative modeling approach for benchmarking and training shallow quantum circuits[J]. NPJ Quantum Information, 2019, 5(1): 45−56
26
Huang Heliang, Du Yuxuan, Gong Ming, et al. Experimental quantum generative adversarial networks for image generation[J]. Physical Review Applied, 2021, 16(2): 024051
27
Silver D, Patel T, Cutler W, et al. MosaiQ: Quantum generative adversarial networks for image generation on NISQ computers[C] //Proc of the IEEE/CVF Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 7030−7039
28
Qu Zhiguo, Shi Wenke, Tiwari P. Quantum conditional generative adversarial network based on patch method for abnormal electrocardiogram generation[J]. Computers in Biology and Medicine, 2023, 166: 107549
29
韩京宇,陈伟,赵静. 基于异常特征模式的心电数据标签清洗方法[J]. 计算机研究与发展,2023,60(11):2594−2610
Han Jingyu, Chen Wei, Zhao Jing. A label cleaning method of ECG data based on abnormality-feature patterns[J]. Journal of Computer Research and Development, 2023, 60(11): 2594−2610 (in Chinese)
30
Arjovsky M, Chintala S, Bottou L. Wasserstein generative adversarial networks[C] // Proc of the 34th Int Conf on Machine Learning. New York: PMLR, 2017: 214−223
31
George M. MIT-BIH arrhythmia database [DB/OL]. [2024-05-01]. https://physionet.org/content/mitdb/1.0.0/
32
Paul A. MIT-BIH ST change database [DB/OL]. [2024-05-01]. https://physionet.org/content/stdb/1.0.0/
33
Taddei A. European ST-T database [DB/OL]. [2024-05-01]. https://physionet.org/content/edb/1.0.0/
34
Greenwald S D. Sudden cardiac death holter database [DB/OL]. [2024-05-01]. https://physionet.org/content/sddb/1.0.0/
35
Borji A. Pros and cons of GAN evaluation measures[J]. Computer Vision and Image Understanding, 2019, 179: 41−65
36
Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. arXiv preprint, arXiv: 1511.06434, 2015
2025年第62卷第7期
114
50
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202440527
  • 接收时间:2024-06-17
  • 首发时间:2026-11-10
  • 出版时间:2025-07-01
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-17
  • 修回日期:2025-02-28
  • 录用日期:2025-04-03
基金
作者信息
    1南京信息工程大学计算机学院、网络空间安全学院 南京 210044
    2南京信息工程大学软件工程学院 南京 210044
    3浙江师范大学计算机科学与技术学院 浙江金华 321004
    4鹏城实验室新型网络研究部 广东深圳 518066

通讯作者:

参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202440527
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏