浏览全部资源
扫码关注微信
纸质出版日期： 2025-02-16 ，
移动端阅览
赵小明， 石培炼， 王丹丹， 付有瑶， 张石清， 方江雄. 2025. 分布感知均值教师网络的半监督医学影像分割. 中国图象图形学报， 30(02):0575-0588
Zhao Xiaoming， Shi Peilian， Wang Dandan， Fu Youyao， Zhang Shiqing， Fang Jiangxiong. 2025. Distribution-aware mean teacher networks for semi-supervised medical image segmentation. Journal of Image and Graphics， 30(02):0575-0588
目的
2
半监督方法旨在通过将未标记数据与标记数据的训练相结合，能够减少对标记数据的依赖，并取得较好的医学图像分割结果。然而，现有的半监督方法通常没有关注到标记数据和未标记数据之间的分布差异所带来的不利影响，尤其是在标记数据比例较低时，可能会严重影响模型的分割性能。
方法
2
提出一种采用分布感知的均值教师网络 （distribution-aware mean teacher network，DAMTN）用于半监督医学影像分割。该方法利用标记数据和未标记数据的分布信息来指导模型的学习，以便在训练阶段使模型对标记和未标记数据的分割结果的分布尽可能相似。该方法采用老师—学生的网络架构，并嵌入了不同的注意力模块，以及分布感知（distribution-aware，DA）模块、完整性监督（integrity supervision，IS）模块和不确定性最小化（uncertainty minimization，UM）模块。
结果
2
在MICCAI 2018（Medical Image Computing and Computer Assisted Intervention Society）左心房分割挑战LA（left atrium）数据集和胰腺CT（computed tomography）数据集上的实验结果表明，该方法使用左心房10%标记数据时，获得的Dice系数、Jaccard指数、HD（Hausdorff distance）和ASD（average surface distance）分别为88.55%、79.62%、9.07和3.08，与基于不确定性的协同均值教师（uncertainty-guided collaborative mean teacher，UCMT）相比，Dice系数和Jaccard指数分别提高了0.42%和0.44%；而使用左心房20%标记数据时，获得的Dice系数、Jaccard指数、HD距离和ASD距离分别为90.55%、82.82%、5.78和1.77，与UCMT相比，Dice系数和Jaccard指数分别提高了0.14%和0.28%。该方法使用胰腺CT 10%标记数据时，获得的Dice系数、Jaccard指数、HD和ASD分别为70.20%、56.36%、15.64和3.57，与基于不确定性的互补一致性学习（uncertainty-guided mutual consistency learning，UG-MCL）相比，Dice系数和Jaccard指数分别提高了0.94%和1.06%；而使用胰腺CT 20%标记数据时，获得的Dice系数、Jaccard指数、HD距离和ASD距离分别为77.89%、64.92%、7.97和1.65，与UG-MCL相比，Dice系数和Jaccard指数分别提高了2.77%和3.34%。在ACDC（automated cardiac diagnosis challenge）数据集上的实验结果也表明了该方法的优越性。
结论
2
提出的方法利用标记数据与未标记数据的分布差异信息，有效提升了半监督医学影像分割性能。尤其在使用较低数量的标记数据时，该方法的分割性能明显优于其他使用的半监督方法。
Objective
2
Medical image segmentation has great potential for application in clinical diagnosis. Although supervised medical image segmentation models can achieve good segmentation results， they rely heavily on pixel-level annotated training data. However， acquiring labeled data is costly and can only be performed by experts. In particular， in the case of 3D medical datasets， the annotation process is more intricate and time-consuming than in 2D datasets. Semi-supervised learning is an effective solution that combines labeled and unlabeled data for improved segmentation results. However， existing semi-supervised methods do not address the performance impact caused by the distribution gap between labeled and unlabeled data， especially when the proportion of labeled data is low.
Method
2
A distribution-aware mean teacher network （DAMTN） is proposed for semi-supervised medical image segmentation to reduce the performance impact caused by the distribution gap between labeled and unlabeled data. This method utilizes the distribution information of both labeled and unlabeled data to guide the learning of the model， aiming to ensure that the segmentation results of labeled and unlabeled data have similar distributions during the training phase. The DAMTN adopts a teacher-student architecture， consisting of a teacher model and a student model. In addition， both the teacher model and the student model’s network architectures are based on the V-N
et design， which includes an encoder and three decoders. The residual connections between the encoder and decoder have been removed. The decoders are differentiated by embedding different attention modules， which are used to introduce perturbations at the model level. These attention methods， including cross-sampling mutual attention （CMA）， position attention （PA）， and channel attention （CA）， are employed to process high-level features. The CMA is used for inter-sample feature interaction and alignment， the PA is employed to handle spatial position information in the feature maps， the CA is utilized to address channel correlations in the feature maps， and the DAMTN comprises three key modules： the distribution-aware （DA） module， integrity supervision （IS）， and uncertainty minimization （UM） module. These three modules compute the losses for the outputs of the student model， and the outputs of the teacher model are used to calculate the consistency loss with the corresponding outputs of the student model. The DA module encourages the model to learn features when the distribution information from labeled and unlabeled data is similar. Each decoder branch is equipped with a dual normalization block consisting of two normalization layers （
<math id="M1"><msub><mrow><mi>B</mi></mrow><mrow><mn mathvariant="normal">0</mn></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73508986&type=
3.21733332
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73508989&type=
3.04800010
and
<math id="M2"><msub><mrow><mi>B</mi></mrow><mrow><mn mathvariant="normal">1</mn></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509000&type=
3.21733332
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509003&type=
3.04800010
） to obtain the distribution information of labeled and unlabeled data.
<math id="M3"><msub><mrow><mi>B</mi></mrow><mrow><mn mathvariant="normal">0</mn></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509006&type=
3.21733332
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509009&type=
3.04800010
handles the unlabeled data， whereas
<math id="M4"><msub><mrow><mi>B</mi></mrow><mrow><mn mathvariant="normal">1</mn></mrow></msub></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509018&type=
3.21733332
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=73509021&type=
3.04800010
handles the labeled data. The IS module supervises the predicted mean of the branches and encourages the model to focus on challenging regions， further reducing the differences between branches. The UM module aims to balance the uncertainty of different attention branches， thereby making the predictions
of all branches more consistent and improving the model’s confidence. The student model incorporates an IS module and a UM module during the training phase to improve the segmentation results， thereby ensuring the consistency and accuracy of the outputs from the three decoders. During the testing phase， the student model takes the average output of the three decoders as its final output. In addition， the parameters of the student model are updated by the optimizer， whereas the parameters of the teacher model are updated using the exponential moving average.
Result
2
Experiments were conducted on the Medical Image Computing and Computer Assisted Intervention Society 2018（MICCAI 2018）—Left Atrial Segmentation Challenge（LA） dataset and the Pancreas CT dataset. In the experiment with 10% labeled data for left atrium segmentation， the Dice coefficient， Jaccard index， Hausdorff distance （HD）， and average surface distance （ASD） were 88.55%， 79.62%， 9.07， and 3.08， respectively. In the experiment with 20% labeled data for the left atrium， the Dice coefficient， Jaccard index， HD， and ASD were 90.55%， 82.82%， 5.78， and 1.77， respectively. For the LA dataset， when trained with 10% labeled data， compared with those of the uncertainty-guided collaborative mean teacher （UCMT）， the Dice coefficient and Jaccard index improved by 0.42% and 0.44%， respectively. For the LA dataset， when trained with 20% labeled data， compared with those of the UCMT， the Dice coefficient and Jaccard index improved by 0.14% and 0.28%， respectively. In the experiment with 10% labeled data for Pancreas CT， the Dice coefficient， Jaccard index， HD， and ASD were 70.20%， 56.36%， 15.64， and 3.57， respectively. In the experiment with 20% labeled data for Pancreas CT， the Dice coefficient， Jaccard index， HD， and ASD were 77.89%， 64.92%， 7.97， and 1.65， respectively. For the Pancreas dataset， when trained with 10% labeled data， compared with uncertainty-guided mutual consistency learning （UG-MCL）， the Dice coefficient and Jaccard index improved by 0.94% and 1.06%， respectively. Additionally， compared with those of the mutual consistency network （MC-Net）， the Dice coefficient and Jaccard index improved by 3.70% and 4.00%， respectively. For the Pancreas dataset， when trained with 20% labeled data， compared with UG-MCL， the Dice coefficient and Jaccard index improved by 2.77% and 3.34%， respectively. Additionally， compared with those of MC-Net， the Dice coefficient and Jaccard index improved by 0.73% and 0.61%， respectively. Additionally， the experimental results on the automated cardiac diagnosis challenge（ACDC） dataset also demonstrate the superiority of this method.
Conclusion
2
The model proposed in this study is based on the teacher-student framework， incorporates attention mechanisms， and leverages the distribution information of both labeled and unlabeled data to constrain the student model. It effectively addresses the distribution gap between labeled and unlabeled data， reducing its impact on performance and improving segmentation results. In particular， when labeled data are scarce， this model outperforms other semi-supervised segmentation methods in terms of segmentation performance.
Bachman P ， Hjelm R D and Buchwalter W . 2019 . Learning representations by maximizing mutual information across views // Proceedings of the 33rd International Conference on Neural Information Processing Systems . Vancouver， Canada ： Curran Associates Inc.： 15535 - 15545
Bernard O ， Lalande A ， Zotti C ， Cervenansky F ， Yang X ， Heng P A ， Cetin I ， Lekadir K ， Camara O ， Ballester M A G ， Sanroma G ， Napel S ， Petersen S ， Tziritas G ， Grinias E ， Khened M ， Kollerathu V A ， Krishnamurthi G ， Rohé M M ， Pennec X ， Sermesant M ， Isensee F ， Jäger P ， Maier-Hein K H ， Full P M ， Wolf I ， Engelhardt S ， Baumgartner C F ， Koch L M ， Wolterink J M ， Išgum I ， Jang Y ， Hong Y ， Patravali J ， Jain S ， Humbert O and Jodoin P M . 2018 . Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis： is the problem solved？ IEEE Transactions on Medical Imaging ， 37 （ 11 ）： 2514 - 2525 ［ DOI： 10.1109/TMI.2018.2837502 http://dx.doi.org/10.1109/TMI.2018.2837502 ］
Chen T ， Kornblith S ， Norouzi M and Hinton G . 2020 . A simple framework for contrastive learning of visual representations // Proceedings of the 37th International Conference on Machine Learning . Virtual Event ： JMLR.org： 1597 - 1607
Chung Y ， Lim C ， Huang C ， Marrouche N and Hamm J . 2023 . FBA-Net： foreground and background aware contrastive learning for semi-supervised atrium segmentation // Proceedings of the 2nd International Workshop on Medical Image Learning with Limited and Noisy Data . Vancouver， Canada ： Springer： 106 - 116 ［ DOI： 10.1007/978-3-031-44917-8_10 http://dx.doi.org/10.1007/978-3-031-44917-8_10 ］
Çiçek Ö ， Abdulkadir A ， Lienkamp S S ， Brox T and Ronneberger O . 2016 . 3D U-Net： learning dense volumetric segmentation from sparse annotation // Proceedings of the 19th International Conference on Medical Image Computing and Computer-Assisted Intervention . Athens， Greece ： Springer： 424 - 432 ［ DOI： 10.1007/978-3-319-46723-8_49 http://dx.doi.org/10.1007/978-3-319-46723-8_49 ］
Fu J ， Liu J ， Tian H J ， Li Y ， Bao Y J ， Fang Z W and Lu H Q . 2019 . Dual attention network for scene segmentation // Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Long Beach， USA ： IEEE： 3141 - 3149 ［ DOI： 10.1109/cvpr.2019.00326 http://dx.doi.org/10.1109/cvpr.2019.00326 ］
Gao S B ， Zhang Z J ， Ma J C ， Li Z H and Zhang S . 2023 . Correlation-aware mutual learning for semi-supervised medical image segmentation // Proceedings of the 26th International Conference on Medical Image Computing and Computer Assisted Intervention . Vancouver， Canada ： Springer： 98 - 108 ［ DOI： 10.1007/978-3-031-43907-0_10 http://dx.doi.org/10.1007/978-3-031-43907-0_10 ］
Huang H M ， Lin L F ， Tong R F ， Hu H J ， Zhang Q W ， Iwamoto Y ， Han X H ， Chen Y W and Wu J . 2020 . UNet 3 +： a full-scale connected UNet for medical image segmentation// Proceedings of 2020 International Conference on Acoustics， Speech and Signal Processing （ICASSP） . Barcelona， Spain ： IEEE： 1055 - 1059 ［ DOI： 10.1109/icassp40776.2020.9053405 http://dx.doi.org/10.1109/icassp40776.2020.9053405 ］
Hung W C ， Tsai Y H ， Liou Y T ， Lin Y Y and Yang M H . 2018 . Adversarial learning for semi-supervised semantic segmentation ［EB/OL］. ［ 2024-04-11 ］. http : /arxiv . org/pdf/1802.07934 .pdf
Kirillov A ， Mintun E ， Ravi N ， Mao H Z ， Rolland C ， Gustafson L ， Xiao T T ， Whitehead S ， Berg A C ， Lo W Y ， Dollr P and Girshick R . 2023 . Segment anything // Proceedings of 2023 IEEE/CVF International Conference on Computer Vision . Paris， France ： IEEE： 3992 - 4003 ［ DOI： 10.1109/ICCV51070.2023.00371 http://dx.doi.org/10.1109/ICCV51070.2023.00371 ］
Li S L ， Zhang C Y and He X M . 2020 . Shape-aware semi-supervised 3D semantic segmentation for medical images // Proceedings of the 23rd International Conference on Medical Image Computing and Computer Assisted Intervention . Lima， Peru ： Springer： 552 - 561 ［ DOI： 10.1007/978-3-030-59710-8_54 http://dx.doi.org/10.1007/978-3-030-59710-8_54 ］
Luo X D ， Chen J N ， Song T and Wang G T . 2021 . Semi-supervised medical image segmentation through dual-task consistency // Proceedings of the 35th AAAI Conference on Artificial Intelligence . Virtually ： AAAI： 8801 - 8809 ［ DOI： 10.1609/aaai.v35i10.17066 http://dx.doi.org/10.1609/aaai.v35i10.17066 ］
Milletari F ， Navab N and Ahmadi S A . 2016 . V-Net： fully convolutional neural networks for volumetric medical image segmentation // Proceedings of the 4th International Conference on 3D Vision （3DV） . Stanford， USA ： IEEE： 565 - 571 ［ DOI： 10.1109/3dv.2016.79 http://dx.doi.org/10.1109/3dv.2016.79 ］
Oktay O ， Schlemper J ， Le Folgoc L ， Lee M ， Heinrich M ， Misawa K ， Mori K ， McDonagh S ， Hammerla N Y ， Kainz B ， Glocker B and Rueckert D . 2018 . Attention U-Net： learning where to look for the pancreas ［EB/OL］. ［ 2024-04-11 ］. https://arxiv.org/pdf/1804.03999.pdf https://arxiv.org/pdf/1804.03999.pdf
Qin X B ， Zhang Z C ， Huang C Y ， Dehghan M ， Zaiane O R and Jagersand M . 2020 . U2-Net： going deeper with nested U-structure for salient object detection . Pattern Recognition ， 106 ： # 107404 ［ DOI： 10.1016/j.patcog.2020.107404 http://dx.doi.org/10.1016/j.patcog.2020.107404 ］
Ronneberger O ， Fischer P and Brox T . 2015 . U-net： convolutional networks for biomedical image segmentation // Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention . Munich， Germany ： Springer： 234 - 241 ［ DOI： 10.1007/978-3-319-24574-4_28 http://dx.doi.org/10.1007/978-3-319-24574-4_28 ］
Shen Z Q ， Cao P ， Yang H ， Liu X L ， Yang J Z and Zaiane O R . 2023 . Co-training with high-confidence pseudo labels for semi-supervised medical image segmentation // Proceedings of the 32nd International Joint Conference on Artificial Intelligence . Macau， China ： Morgan Kaufmann： 4199 - 4207 ［ DOI： 10.24963/ijcai.2023/467 http://dx.doi.org/10.24963/ijcai.2023/467 ］
Shi Y Y ， Zhang Y C and Wang S S . 2023 . Competitive ensembling teacher-student framework for semi-supervised left atrium MRI segmentation // Proceedings of 2023 IEEE International Conference on Bioinformatics and Biomedicine （BIBM） . Istanbul， Turkey ： IEEE： 2231 - 2234 ［ DOI： 10.1109/bibm58861.2023.10385958 http://dx.doi.org/10.1109/bibm58861.2023.10385958 ］
Sohn K ， Berthelot D ， Li C L ， Zhang Z Z ， Carlini N ， Cubuk E D ， Kurakin A ， Zhang H and Raffel C . 2020 . FixMatch： simplifying semi-supervised learning with consistency and confidence // Proceedings of the 34th International Conference on Neural Information Processing Systems . Vancouver， Canada ： Curran Associates Inc.： 596 - 608
Su Y X and Xiao Z Y . 2023 . Semi-supervised dual-task cross-consistency constraint network for 3D cardiac MRI segmentation . Journal of Image and Graphics ， 28 （ 4 ）： 1198 - 1211
苏逸欣 ， 肖志勇 . 2023 . 面向3维心脏MRI分割的半监督双任务交叉一致性约束网络 . 中国图象图形学报 ， 28 （ 4 ）： 1198 - 1211 ［ DOI： 10.11834/jig.211019 http://dx.doi.org/10.11834/jig.211019 ］
Tarvainen A and Valpola H . 2017 . Mean teachers are better role models： weight-averaged consistency targets improve semi-supervised deep learning results // Proceedings of the 31st International Conference on Neural Information Processing Systems . Long Beach， USA ： Curran Associates Inc.： 1195 - 1204
Vu T-H ， Jain H ， Bucher M ， Cord M ， Pérez P . 2019 . Advent： adversarial entropy minimization for domain adaptation in semantic segmentation // Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Long Beach， USA ： IEEE： 2517 - 2526 ［ DOI： 10.1109/cvpr.2019.00262 http://dx.doi.org/10.1109/cvpr.2019.00262 ］
Wang K P ， Zhan B ， Zu C ， Wu X ， Zhou J L ， Zhou L P and Wang Y . 2021 . Tripled-uncertainty guided mean teacher model for semi-supervised medical image segmentation // Proceedings of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention . Strasbourg， France ： Springer： 450 - 460 ［ DOI： 10.1007/978-3-030-87196-3_42 http://dx.doi.org/10.1007/978-3-030-87196-3_42 ］
Woo S ， Park J ， Lee J Y and Kweon I S . 2018 . CBAM： convolutional block attention module // Proceedings of the 15th European Conference on Computer Vision . Munich， Germany ： Springer： 3 - 19 ［ DOI： 10.1007/978-3-030-01234-2_1 http://dx.doi.org/10.1007/978-3-030-01234-2_1 ］
Wu Y C ， Ge Z Y ， Zhang D H ， Xu M F ， Zhang L ， Xia Y and Cai J F . 2022a . Mutual consistency learning for semi-supervised medical image segmentation . Medical Image Analysis ， 81 ： # 102530 ［ DOI： 10.1016/j.media.2022.102530 http://dx.doi.org/10.1016/j.media.2022.102530 ］
Wu Y C ， Wu Z H ， Wu Q Y ， Ge Z Y and Cai J F . 2022b . Exploring smoothness and class-separation for semi-supervised medical image segmentation // Proceedings of the 25th International Conference on Medical Image Computing and Computer Assisted Intervention . Singapore， Singapore ： Springer： 34 - 43 ［ DOI： 10.1007/978-3-031-16443-9_4 http://dx.doi.org/10.1007/978-3-031-16443-9_4 ］
Wu Y C ， Xu M F ， Ge Z Y ， Cai J F and Zhang L . 2021 . Semi-supervised left atrium segmentation with mutual consistency training // Proceedings of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention . Strasbourg， France ： Springer： 297 - 306 ［ DOI： 10.1007/978-3-030-87196-3_28 http://dx.doi.org/10.1007/978-3-030-87196-3_28 ］
Yu L Q ， Wang S J ， Li X M ， Fu C W and Heng P A . 2019 . Uncertainty-aware self-ensembling model for semi-supervised 3D left atrium segmentation // Proceedings of the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention . Shenzhen， China ： Springer： 605 - 613 ［ DOI： 10.1007/978-3-030-32245-8_67 http://dx.doi.org/10.1007/978-3-030-32245-8_67 ］
Zhang Y C ， Jiao R S ， Liao Q C ， Li D Y and Zhang J C . 2023 . Uncertainty-guided mutual consistency learning for semi-supervised medical image segmentation . Artificial Intelligence in Medicine ， 138 ： # 102476 ［ DOI： 10.1016/j.artmed.2022.102476 http://dx.doi.org/10.1016/j.artmed.2022.102476 ］
Zhang Y C ， Yang J ， Liu Y C ， Cheng Y and Qi Y . 2024 . SemiSAM： enhancing semi-supervised medical image segmentation via SAM-assisted consistency regularization ［EB/OL］. ［ 2024-05-18 ］. https://arxiv.org/pdf/2312.06316.pdf https://arxiv.org/pdf/2312.06316.pdf
Zhao X M ， Liao Y H ， Zhang S Q ， Fang J X ， He X X ， Wang G Y and Lu H S . 2024 . Method of classifying benign and malignant breast tumors by DCE-MRI incorporating local and global features . Journal of Image and Graphics ， 29 （ 1 ）： 256 - 267
赵小明 ， 廖越辉 ， 张石清 ， 方江雄 ， 何遐遐 ， 汪国余 ， 卢洪胜 . 2024 . 融合局部与全局特征的DCE-MRI乳腺肿瘤良恶分类 . 中国图象图形学报 ， 29 （ 1 ）： 256 - 267 ［ DOI： 10.11834/jig.230092 http://dx.doi.org/10.11834/jig.230092 ］
Zhong L F ， Liao X ， Zhang S T and Wang G T . 2023 . Semi-supervised pathological image segmentation via cross distillation of multiple attentions // Proceedings of the 26th International Conference on Medical Image Computing and Computer Assisted Intervention . Vancouver， Canada ： Springer： 570 - 579 ［ DOI： 10.1007/978-3-031-43987-2_55 http://dx.doi.org/10.1007/978-3-031-43987-2_55 ］
Zhou X G ， Li Z Q and Tong T . 2022a . DTSC-Net： semi-supervised 3D biomedical image segmentation through dual-teacher simplified consistency // Proceedings of 2022 IEEE International Conference on Bioinformatics and Biomedicine （BIBM） . Las Vegas， USA ： IEEE： 1429 - 1434 ［ DOI： 10.1109/bibm55620.2022.9995371 http://dx.doi.org/10.1109/bibm55620.2022.9995371 ］
Zhou Z Q ， Qi L ， Yang X ， Ni D and Shi Y H . 2022b . Generalizable cross-modality medical image segmentation via style augmentation and dual normalization // Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition . New Orleans， USA ： IEEE： 20824 - 20833 ［ DOI： 10.1109/cvpr52688.2022.02019 http://dx.doi.org/10.1109/cvpr52688.2022.02019 ］
Zhou Z W ， Rahman Siddiquee M M ， Tajbakhsh N and Liang J M . 2018 . UNet++： a nested U-net architecture for medical image segmentation // Proceedings of the 4th International Workshop on Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support . Granada， Spain ： Springer： 3 - 11 ［ DOI： 10.1007/978-3-030-00889-5_1 http://dx.doi.org/10.1007/978-3-030-00889-5_1 ］
相关作者
相关机构
京公网安备11010802024621