基于标签噪声对比学习的肺癌淋巴结转移鉴别
作者:
基金项目:

甘肃省高等学校青年博士基金 (2022QB-118)


Identification of Lung Cancer Lymph Node Metastasis Based on Label Noise Contrastive Learning
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [33]
  • |
  • 相似文献 [20]
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    基于深度学习的人工智能诊断模型严重依赖于高质量地详尽注释数据进行算法训练, 但受到标签噪声信息的影响. 为了增强模型的鲁棒性并防止有噪声的标签记忆, 提出了一种标签噪声样本选择 (noise label sample selection, NLSS)模型来充分挖掘噪声样本的隐藏信息, 减轻模型过拟合问题. 首先, 通过将混合增强图像作为输入, 提取图像分布式特征表示; 其次, 引入对比损失函数以及比较样本预测标签分布与其真实标签分布的相似性来评估样本, 进行样本选择; 最后, 通过标签重分配模块的伪标签提升策略在样本选择的基础上重新纠正噪声标签的监督信息. 以非小细胞肺癌 (non-small cell lung cancer, NSCLC)患者的 PET/CT 数据集为例进行实验, 结果表明提出的模型均比对比模型有一定的提升, 可降低淋巴结转移状态诊断中标签噪声的干扰.

    Abstract:

    The AI diagnostic model based on deep learning relies heavily on high-quality detailed annotated data for algorithm training, but is affected by label noise information. To enhance the robustness of the model and prevent noisy label memory, a noise label sample selection (NLSS) model is proposed to fully mine the hidden information of noise samples and alleviate model overfitting. Firstly, distributed feature representations of the image are extracted by taking hybrid enhanced images as input. Secondly, the contrasive loss function is introduced to compare the similarity between the predicted label distribution of the sample and the real label distribution for sample evaluation and selection. Finally, based on sample selection, supervised information of the noisy label is re-corrected by the pseudo-label promotion strategy of the label redistribution module. Taking the PET/CT dataset of non-small cell lung cancer (NSCLC) patients as an example, results show that the proposed models outperform comparison models, reducing the interference of label noise in the diagnosis of lymph node metastasis.

    参考文献
    [1] Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 2021, 71(3): 209–249.
    [2] 陈武飞, 马庄宣, 李铭. 影像组学在非小细胞肺癌诊疗中的应用进展. 中国医学影像学杂志, 2023, 31(7): 781–784.
    [3] Hendrycks D, Mazeika M, Wilson D, et al. Using trusted data to train deep networks on labels corrupted by severe noise. Proceedings of the 32nd International Conference on Neural Information Processing Systems (NIPS 2018). Montreal: Curran Associates Inc., 2018. 10477–10486.
    [4] Jiang L, Zhou ZY, Leung T, et al. MentorNet: Learning data-driven curriculum for very deep neural networks on corrupted labels. Proceedings of the 35th International Conference on Machine Learning. Stockholm: PMLR, 2018. 2304–2313.
    [5] Liu JR, Li RR, Sun C. Co-correcting: Noise-tolerant medical image classification via mutual label correction. IEEE Transactions on Medical Imaging, 2021, 40(12): 3580–3592.
    [6] Zhang ZL, Sabuncu MR. Generalized cross entropy loss for training deep neural networks with noisy labels. Proceedings of the 32nd International Conference on Neural Information Processing Systems (NIPS 2018). Montreal: Curran Associates Inc., 2018. 8792–8802.
    [7] Xu YL, Cao P, Kong YQ, et al. LDMI: A novel information-theoretic loss function for training deep nets robust to label noise. Proceedings of the 33rd International Conference on Neural Information Processing Systems (NeurIPS). Vancouver: Curran Associates Inc., 2019. 559.
    [8] Ma XJ, Huang HX, Wang YS, et al. Normalized loss functions for deep learning with noisy labels. Proceedings of the 37th International Conference on Machine Learning. PMLR, 2020. 6543–6553.
    [9] Xue C, Dou Q, Shi XY, et al. Robust learning at noisy labeled medical images: Applied to skin lesion classification. Proceedings of the 16th IEEE International Symposium on Biomedical Imaging (ISBI). Venice: IEEE, 2019. 1280–1283.
    [10] Min SB, Chen XJ, Zha ZJ, et al. A two-stream mutual attention network for semi-supervised biomedical segmentation with noisy labels. Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Honolulu: AAAI, 2019. 4578–4585.
    [11] Xing XH, Chen Z, Gao ZF, et al. Gradient and feature conformity-steered medical image classification with noisy labels. Proceedings of the 26th International Conference Medical Image Computing and Computer Assisted Intervention. Vancouver: Springer, 2023. 75–84.
    [12] Chen T, Kornblith S, Norouzi M, et al. A simple framework for contrastive learning of visual representations. Proceedings of the 37th International Conference on Machine Learning (ICML). PMLR, 2020. 1597–1607.
    [13] 李俊哲, 曹国. 基于对比学习的标签带噪图像分类. 计算机系统应用, 2023, 32(12): 104–111.
    [14] Li SK, Xia XB, Ge SM, et al. Selective-supervised contrastive learning with noisy labels. Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022. 316–325.
    [15] Xu J, Huang Y, Cheng MM, et al. Noisy-as-clean: Learning self-supervised denoising from corrupted image. IEEE Transactions on Image Processing, 2020, 29: 9316–9329.
    [16] He KM, Zhang XY, Ren SQ, et al. Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016. 770–778.
    [17] Song H, Kim M, Park D, et al. Learning from noisy labels with deep neural networks: A survey. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(11): 8135–8153.
    [18] Han B, Yao QM, Yu XR, et al. Co-teaching: Robust training of deep neural networks with extremely noisy labels. Proceedings of the 32nd International Conference on Neural Information Processing Systems (NIPS 2018). Montreal: Curran Associates Inc., 2018. 8536–8546.
    [19] Yu XR, Han B, Yao JC, et al. How does disagreement help generalization against label corruption? Proceedings of the 36th International Conference on Machine Learning. Long Beach: PMLR, 2019. 7164–7173.
    [20] Wei HX, Feng L, Chen XY, et al. Combating noisy labels by agreement: A joint training method with co-regularization. Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 2020. 13723–13732.
    [21] Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016. 2818–2826.
    [22] 李晓艳, 张子刚, 张逸石, 等. 一种基于KL散度和类分离策略的特征选择算法. 计算机科学, 2012, 39(12): 224–227.
    [23] Le PB, Nguyen ZT. ROC curves, loss functions, and distorted probabilities in binary classification. Mathematics, 2022, 10(9): 1410.
    [24] 程磊, 吴晓富, 张索非. 数据集类别不平衡性对迁移学习的影响分析. 信号处理, 2020, 36(1): 110–117.
    [25] Zeng H, Cui XH. SimCLRT: A simple framework for contrastive learning of rumor tracking. Engineering Applications of Artificial Intelligence, 2022, 110: 104757.
    [26] Yi L, Liu S, She Q, et al. On learning contrastive representations for learning with noisy labels. Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans: IEEE, 2022. 16661–16670.
    [27] He KM, Fan HQ, Wu YX, et al. Momentum contrast for unsupervised visual representation learning. Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 2020. 9726–9735.
    [28] Paszke A, Gross S, Massa F, et al. PyTorch: An imperative style, high-performance deep learning library. Proceedings of the 33rd International Conference on Neural Information Processing Systems. Vancouver: Curran Associates Inc., 2019. 721.
    [29] 常禧龙, 梁琨, 李文涛. 深度学习优化器进展综述. 计算机工程与应用, 2024, 60(7): 1–12.
    [30] 张剑飞, 王真, 崔文升, 等. 基于交叉验证和神经网络融合的医学数据分类. 齐齐哈尔大学学报(自然科学版), 2019, 35(4): 1–5.
    [31] Aung N, Kechadi T, Chen LM, et al. IP-UNet: Intensity projection UNet architecture for 3D medical volume segmentation. arXiv:2308.12761, 2023.
    [32] Yao YZ, Sun ZR, Zhang CY, et al. Jo-SRC: A contrastive approach for combating noisy labels. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021. 5188–5197.
    [33] Liu FB, Chen YH, Wang C, et al. Asymmetric co-teaching with multi-view consensus for noisy label learning. arXiv:2301.01143, 2023.
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

祁婧,李子荣,刘秀婷,马露,陈俊豪.基于标签噪声对比学习的肺癌淋巴结转移鉴别.计算机系统应用,2024,33(12):177-184

复制
分享
文章指标
  • 点击次数:88
  • 下载次数: 513
  • HTML阅读次数: 376
  • 引用次数: 0
历史
  • 收稿日期:2024-05-22
  • 最后修改日期:2024-06-17
  • 在线发布日期: 2024-10-25
文章二维码
您是第11197721位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号