基于自监督和半监督学习的医学影像分割方法

胶质瘤 MR 图像和支气管 CT 图像的准确分割是胶质瘤和肺部诊治的重要前提，然而模态缺失会导致常见的胶质瘤自动分割方法精度严重下降，而支气管远端小分支在现有方法下难以被准确分割。提出了两个创新的深度学习模型，旨在通过先进的自监督学习和半监督学习技术，解决上述问题，提高多种模态医学影像分割的精度和效率。对于模态缺失的胶质瘤分割，开发了一个基于图像遮蔽重建的自监督学习模型。该模型首先通过遮蔽重建图像的任务学习每种模态丰富的特征表示，使得模型在缺乏完整模态信息的情况下，仍能有效地识别和分割胶质瘤所有区域。在微调模型的分割胶质瘤训练阶段，不仅在输入时模拟了模态缺失的情况，还在模型中加入了一个基于滑动平均的自蒸馏模块。该模块约束模型对于任一病例不同模态缺失的输入提取出具有语义一致性的特征，从而增强模型在面对模态缺失时的强健性。针对支气管小分支难以准确分割的问题，提出了一个基于双扰动一致性的半监督学习模型。借助教师模型生成伪标签的半监督学习方式，弥补了支气管有标注数据集过小的缺陷。用包含伪标签数据和有标注数据的混合数据集训练学生模型时采用了双扰动一致性学习，两种扰动机制分别为基于快速傅里叶变换的图像扰动，用于模拟图像强度的各种变化；和基于随机丢弃特征通道的特征扰动，旨在增强网络对特征表示的泛化能力。双扰动一致性学习有效地指导模型学习到更具判别力的特征，模型泛化性更强，显著改善了对支气管小分支的分割性能。在公开数据集上的实验结果证明，提出的方法在这两种医学影像分割任务中取得了优于现有技术的性能。通过这两项研究，不仅解决了胶质瘤和支气管分割的问题，还展示了自监督学习和半监督学习在处理多种模态的医学影像分割任务中的潜力和有效性。这些成果不仅提升了分割任务的精度，也为处理模态缺失和增强模型泛化能力提供了新的视角和技术路线。

To address the significant challenges in accurate segmentation of glioma MR imagesand bronchial CT images, especially under conditions of missing modalities and difficul ties in segmenting small bronchial branches, two novel deep learning models that leverageadvanced self-supervised and semi-supervised learning techniques are introduced to en hance segmentation accuracy and efficiency.A self-supervised learning model focuses on glioma segmentation by learning richfeature representations through image mask reconstruction tasks, allowing for effec tive glioma identification despite missing modalities. This model incorporates a self distillation module based on moving averages to ensure high semantic consistency acrossdifferent modality absences, thereby improving robustness.To overcome the challenge of segmenting small airway branches, a semi-supervisedmodel utilizes dual-disturbance consistency, combining image disturbances based on fastFourier transform with feature disturbances through random channel dropout. This ap proach facilitates the learning of more discriminative features and improves the model’sgeneralization, leading to superior segmentation performance.Experimental validation on public datasets has shown these models to surpass exist ing segmentation methods for both tasks, demonstrating the potential of self-supervisedand semi-supervised learning in medical image segmentation. These contributions notonly solve specific segmentation problems but also advance the methodology for han dling missing modalities and enhancing model generalization.

[1] WADHWA A, BHARDWAJ A, VERMA V S. A review on brain tumor segmentation of MRI images[J]. Magnetic resonance imaging, 2019, 61: 247-259.
[2] TIWARI A, SRIVASTAVA S, PANT M. Brain tumor segmentation and classification from magnetic resonance images: Review of selected methods from 2014 to 2019[J]. Pattern recognitionletters, 2020, 131: 244-260.
[3] MILLER K D, OSTROM Q T, KRUCHKO C, et al. Brain and other central nervous system tumor statistics, 2021[J]. CA: a cancer journal for clinicians, 2021, 71(5): 381-406.
[4] DAS S, NAYAK G K, SABA L, et al. An artificial intelligence framework and its bias for brain tumor segmentation: A narrative review[J]. Computers in biology and medicine, 2022, 143: 105273.
[5] RAO C S, KARUNAKARA K. A comprehensive review on brain tumor segmentation and classification of MRI images[J]. Multimedia Tools and Applications, 2021, 80(12): 17611-17643.
[6] MENZE B H, JAKAB A, BAUER S, et al. The multimodal brain tumor image segmentation benchmark (BRATS)[J]. IEEE transactions on medical imaging, 2014, 34(10): 1993-2024.
[7] AGRAVAT R R, RAVAL M S. A survey and analysis on automated glioma brain tumor segmentation and overall patient survival prediction[J]. Archives of Computational Methods in Engineering, 2021, 28(5): 4117-4152.
[8] ZHANG W, WU Y, YANG B, et al. Overview of multi-modal brain tumor mr image segmentation[C]//Healthcare: Vol. 9. MDPI, 2021: 1051.
[9] WANG C, XU J, YANG L, et al. Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China Pulmonary Health [CPH] study): a national cross-sectional study[J]. The Lancet, 2018, 391(10131): 1706-1717.
[10] NAKANO Y, MURO S, SAKAI H, et al. Computed tomographic measurements of airway dimensions and emphysema in smokers: correlation with lung function[J]. American journal of respiratory and critical care medicine, 2000, 162(3): 1102-1108.
[11] BERGER P, PEROT V, DESBARATS P, et al. Airway wall thickness in cigarette smokers: quantitative thin-section CT assessment[J]. Radiology, 2005, 235(3): 1055-1064.
[12] UKIL S, SONKA M, REINHARDT J M. Automatic segmentation of pulmonary fissures in X-ray CT images using anatomic guidance[C]//Medical Imaging 2006: Image Processing: Vol.6144. SPIE, 2006: 213-223.
[13] MORI K, NAKADA Y, KITASAKA T, et al. Lung lobe and segmental lobe extraction from 3D chest CT datasets based on figure decomposition and Voronoi division[C]//Medical Imaging 2008: Image Processing: Vol. 6914. SPIE, 2008: 1452-1463.
[14] DAVIDSON W, BAI T R. Lung structural changes in chronic obstructive pulmonary diseases[J]. Current Drug Targets-Inflammation & Allergy, 2005, 4(6): 643-649.
[15] VAN RIKXOORT E M, VAN GINNEKEN B. Automated segmentation of pulmonary structures in thoracic computed tomography scans: a review[J]. Physics in Medicine & Biology, 2013, 58(17): R187.
[16] SLUIMER I, SCHILHAM A, PROKOP M, et al. Computer analysis of computed tomography scans of the lung: a survey[J]. IEEE transactions on medical imaging, 2006, 25(4): 385-405.
[17] LI X, TAN W, LIU P, et al. Classification of COVID-19 chest CT images based on ensemble deep learning[J]. Journal of healthcare engineering, 2021, 2021.
[18] TAN W, YUAN Y, CHEN A, et al. An approach for pulmonary vascular extraction from chest CT images[J]. Journal of healthcare engineering, 2019, 2019.
[19] PETERSEN J, NIELSEN M, LO P, et al. Optimal surface segmentation using flow lines to quantify airway abnormalities in chronic obstructive pulmonary disease[J]. Medical image analysis, 2014, 18(3): 531-541.
[20] YU K C, GIBBS J D, GRAHAM M W, et al. Image-based reporting for bronchoscopy[J]. Journal of digital imaging, 2010, 23: 39-50.
[21] TAN W, LIU P, LI X, et al. Segmentation of lung airways based on deep learning methods[J]. IET Image Processing, 2022, 16(5): 1444-1456.
[22] KER J, WANG L, RAO J, et al. Deep learning applications in medical image analysis[J]. Ieee Access, 2017, 6: 9375-9389.
[23] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 3431-3440.
[24] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[J]. Advances in neural information processing systems, 2012, 25.
[25] RONNEBERGER O, FISCHER P, BROX T. U-net: Convolutional networks for biomedical image segmentation[C]//Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18. Springer, 2015: 234-241.
[26] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Semantic image segmentation with deep convolutional nets and fully connected crfs[A]. 2014.
[27] ZHOU Z, RAHMAN SIDDIQUEE M M, TAJBAKHSH N, et al. Unet++: A nested u-net architecture for medical image segmentation[C]//Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4. Springer, 2018: 3-11.
[28] CHEN J, LU Y, YU Q, et al. Transunet: Transformers make strong encoders for medical image segmentation[A]. 2021.
[29] CAO H, WANG Y, CHEN J, et al. Swin-unet: Unet-like pure transformer for medical image segmentation[C]//European conference on computer vision. Springer, 2022: 205-218.
[30] KIRILLOV A, MINTUN E, RAVI N, et al. Segment anything[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2023: 4015-4026.
[31] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 40(4): 834-848.
[32] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[A]. 2017.
[33] CHEN L C, ZHU Y, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 801-818.
[34] YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[A]. 2015.
[35] LI C, TAN Y, CHEN W, et al. Attention unet++: A nested attention-aware u-net for liver ct image segmentation[C]//2020 IEEE international conference on image processing (ICIP). IEEE, 2020: 345-349.
[36] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
[37] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//European conference on computer vision. Springer, 2020: 213-229.
[38] LIU Z, LIN Y, CAO Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2021: 10012-10022.
[39] DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[A]. 2020.
[40] HE K, CHEN X, XIE S, et al. Masked autoencoders are scalable vision learners[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2022: 16000-16009.
[41] RADFORD A, KIM J W, HALLACY C, et al. Learning transferable visual models from natural language supervision[C]//International conference on machine learning. PMLR, 2021: 8748-8763.
[42] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30.
[43] ZHOU T, RUAN S, HU H. A literature survey of MR-based brain tumor segmentation with missing modalities[J]. Computerized Medical Imaging and Graphics, 2023, 104: 102167.
[44] ISLAM M, WIJETHILAKE N, REN H. Glioblastoma multiforme prognosis: MRI missing modality generation, segmentation and radiogenomic survival prediction[J]. Computerized Medical Imaging and Graphics, 2021, 91: 101906.
[45] MIRZA M, OSINDERO S. Conditional generative adversarial nets[A]. 2014.
[46] ZHOU T, CANU S, VERA P, et al. Feature-enhanced generation and multi-modality fusion based deep neural network for brain tumor segmentation with missing MR modalities[J]. Neurocomputing, 2021, 466: 102-112.
[47] CHEN C, DOU Q, JIN Y, et al. Robust multimodal brain tumor segmentation via feature disentanglement and gated fusion[C]//Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part III 22. Springer, 2019: 447-456.
[48] HAVAEI M, GUIZARD N, CHAPADOS N, et al. Hemis: Hetero-modal image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016: 19th International Conference, Athens, Greece, October 17-21, 2016, Proceedings, Part II 19. Springer,2016: 469-477.
[49] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[50] CHARTSIAS A, JOYCE T, GIUFFRIDA M V, et al. Multimodal MR synthesis via modalityinvariant latent representation[J]. IEEE transactions on medical imaging, 2017, 37(3): 803-814.
[51] DING Y, YU X, YANG Y. RFNet: Region-aware fusion network for incomplete multi-modal brain tumor segmentation[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2021: 3975-3984.
[52] AZAD R, KHOSRAVI N, MERHOF D. SMU-Net: Style matching U-Net for brain tumor segmentation with missing modalities[C]//International Conference on Medical Imaging with Deep Learning. PMLR, 2022: 48-62.
[53] ZHANG Y, HE N, YANG J, et al. mmformer: Multimodal medical transformer for incomplete multimodal learning of brain tumor segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 2022: 107-117.
[54] HINTON G, VINYALS O, DEAN J. Distilling the knowledge in a neural network[A]. 2015.
[55] HU M, MAILLARD M, ZHANG Y, et al. Knowledge distillation from multi-modal to mono-modal segmentation networks[C]//Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part I 23. Springer, 2020: 772-781.
[56] LOPEZ-PAZ D, BOTTOU L, SCHÖLKOPF B, et al. Unifying distillation and privileged information[A]. 2015.
[57] WANG Y, ZHANG Y, LIU Y, et al. ACN: adversarial co-training network for brain tumor segmentation with missing modalities[C]//Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part VII 24. Springer, 2021: 410-420.
[58] SHEN Y, GAO M. Brain tumor segmentation on MRI with missing modalities[C]//Information Processing in Medical Imaging: 26th International Conference, IPMI 2019, Hong Kong, China, June 2–7, 2019, Proceedings 26. Springer, 2019: 417-428.
[59] LIU H, WEI D, LU D, et al. M3AE: Multimodal representation learning for brain tumor segmentation with missing modalities[C]//Proceedings of the AAAI Conference on Artificial Intelligence: Vol. 37. 2023: 1657-1665.
[60] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[J]. Advances in neural information processing systems, 2014, 27.
[61] GOU J, YU B, MAYBANK S J, et al. Knowledge distillation: A survey[J]. International Journal of Computer Vision, 2021, 129(6): 1789-1819.
[62] ÇIÇEK Ö, ABDULKADIR A, LIENKAMP S S, et al. 3D U-Net: learning dense volumetric segmentation from sparse annotation[C]//Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016: 19th International Conference, Athens, Greece, October 17-21, 2016, Proceedings, Part II 19. Springer, 2016: 424-432.
[63] QIN Y, CHEN M, ZHENG H, et al. AirwayNet: A Voxel-Connectivity Aware Approach for Accurate Airway Segmentation Using Convolutional Neural Networks[C]//SHEN D, LIU T, PETERS T M, et al. Medical Image Computing and Computer Assisted Intervention – MICCAI 2019. Cham: Springer International Publishing, 2019: 212-220.
[64] JUAREZ A G, TIDDENS H A W M, DE BRUIJNE M. Automatic Airway Segmentation in chest CT using Convolutional Neural Networks[J/OL]. CoRR, 2018, abs/1808.04576. http://arxiv.org/abs/1808.04576.
[65] HUANG X, CHOI K S, LIANG S, et al. Frequency Domain Channel-wise Attack to CNN Classifiers in Motor Imagery Brain-Computer Interfaces[J/OL]. IEEE Transactions on Biomedical Engineering, 2023: 1-12. DOI: 10.1109/TBME.2023.3344295.
[66] DING G, WU Y, JIN J, et al. Towards A Relation Extractor Nested U-Architecture for Accurate Pulmonary Airway Segmentation in CT Images[C/OL]//ICCCV ’22: Proceedings of the 5th International Conference on Control and Computer Vision. New York, NY, USA: Association for Computing Machinery, 2022: 29–35. https://doi.org/10.1145/3561613.3561618.
[67] WU Y, ZHANG M, YU W, et al. LTSP: long-term slice propagation for accurate airway segmentation[J/OL]. International Journal of Computer Assisted Radiology and Surgery, 2022, 17(5): 857-865. https://doi.org/10.1007/s11548-022-02582-7.
[68] ZHENG H, QIN Y, GU Y, et al. Alleviating Class-Wise Gradient Imbalance for Pulmonary Airway Segmentation[J/OL]. IEEE Transactions on Medical Imaging, 2021, 40(9): 2452-2462. DOI: 10.1109/TMI.2021.3078828.
[69] ZHENG H, QIN Y, GU Y, et al. Refined Local-Imbalance-Based Weight for Airway Segmentation in CT[C/OL]//Medical Image Computing and Computer Assisted Intervention – MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part I. Berlin, Heidelberg: Springer-Verlag, 2021: 410–419. https://doi.org/10.1007/978-3-030-87193-2_39. DOI: 10.1007/978-3-030-87193-2_39.
[70] GARCIA-UCEDA A, SELVAN R, SAGHIR Z, et al. Automatic airway segmentation from computed tomography using robust and efficient 3-D convolutional neural networks[J/OL]. Scientific Reports, 2021, 11(1): 16001. https://doi.org/10.1038/s41598-021-95364-1.
[71] WANG L, GUO D, WANG G, et al. Annotation-Efficient Learning for Medical Image Segmentation Based on Noisy Pseudo Labels and Adversarial Learning[J/OL]. IEEE Transactions on Medical Imaging, 2021, 40(10): 2795-2807. DOI: 10.1109/TMI.2020.3047807.
[72] YANG L, QI L, FENG L, et al. Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation: abs/2208.09910[A]. 2022.
[73] LUO X, HU M, SONG T, et al. Semi-Supervised Medical Image Segmentation via Cross Teaching between CNN and Transformer[C/OL]//KONUKOGLU E, MENZE B, VENKATARAMAN A, et al. Proceedings of Machine Learning Research: Vol. 172 Proceedings of The 5th International Conference on Medical Imaging with Deep Learning. PMLR, 2022: 820-833. https://proceedings.mlr.press/v172/luo22b.html.
[74] BASAK H, BHATTACHARYA R, HUSSAIN R, et al. An Embarrassingly Simple Consistency Regularization Method for Semi-Supervised Medical Image Segmentation: abs/2202.00677[A]. 2022.
[75] YANG L, ZHUO W, QI L, et al. St++: Make self-training work better for semi-supervised semantic segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 4268-4277.
[76] DEVLIN J, CHANG M W, LEE K, et al. Bert: Pre-training of deep bidirectional transformers for language understanding[A]. 2018.
[77] XIE Z, ZHANG Z, CAO Y, et al. Simmim: A simple framework for masked image modeling[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2022: 9653-9663.
[78] WOO S, DEBNATH S, HU R, et al. Convnext v2: Co-designing and scaling convnets with masked autoencoders[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023: 16133-16142.
[79] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.
[80] WU Y, HE K. Group normalization[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 3-19.
[81] PASZKE A, GROSS S, MASSA F, et al. Pytorch: An imperative style, high-performance deep learning library[J]. Advances in neural information processing systems, 2019, 32.
[82] KINGMA D P, BA J. Adam: A method for stochastic optimization[A]. 2014.
[83] LOSHCHILOV I, HUTTER F. Sgdr: Stochastic gradient descent with warm restarts[A]. 2016.
[84] ARMATO III S G, MCLENNAN G, BIDAUT L, et al. The lung image database consortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans[J]. Medical physics, 2011, 38(2): 915-931.
[85] LO P, VAN GINNEKEN B, REINHARDT J M, et al. Extraction of airways from CT (EXACT’09)[J]. IEEE Transactions on Medical Imaging, 2012, 31(11): 2093-2107.
[86] GEIRHOS R, TEMME C R, RAUBER J, et al. ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness[A]. 2019.
[87] SUN J, DARBEHANI F, ZAIDI M, et al. Saunet: Shape attentive u-net for interpretable medical image segmentation[C]//Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part IV 23. Springer, 2020: 797-806.
[88] QIN Y, ZHENG H, GU Y, et al. Learning Tubule-Sensitive CNNs for Pulmonary Airway and Artery-Vein Segmentation in CT[J/OL]. IEEE Transactions on Medical Imaging, 2021, 40(6): 1603-1617. DOI: 10.1109/TMI.2021.3062280.
[89] YUSHKEVICH P A, PIVEN J, HAZLETT H C, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability[J]. Neuroimage, 2006, 31(3): 1116-1128.
[90] MILLETARI F, NAVAB N, AHMADI S A. V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation[C/OL]//2016 Fourth International Conference on 3D Vision (3DV). 2016: 565-571. DOI: 10.1109/3DV.2016.79.
[91] TARVAINEN A, VALPOLA H. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results[J]. Advances in neural information processing systems, 2017, 30.
[92] WANG C, HAYASHI Y, ODA M, et al. Tubular Structure Segmentation Using Spatial Fully Connected Network with Radial Distance Loss for 3D Medical Images[C/OL]//Medical Image Computing and Computer Assisted Intervention –MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part VI. Berlin, Heidelberg: Springer-Verlag, 2019: 348–356. https://doi.org/10.1007/978-3-030-32226-7_39. DOI: 10.1007/978-3-030-32226-7_39.
[93] GARCIA-UCEDA JUAREZ A, SELVAN R, SAGHIR Z, et al. A Joint 3D UNet-Graph Neural Network-Based Method for Airway Segmentation from Chest CTs[C/OL]//Machine Learning in Medical Imaging: 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 13, 2019, Proceedings. Berlin, Heidelberg: Springer-Verlag, 2019: 583–591. https://doi.org/10.1007/978-3-030-32692-0_67. DOI: 10.1007/978-3-030-32692-0_67.
[94] JIN D, XU Z, HARRISON A P, et al. 3D Convolutional Neural Networks with Graph Refinement for Airway Segmentation Using Incomplete Data Labels[C]//MLMI@MICCAI. 2017.