Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation

Although having achieved great success in medical image segmentation, deep learning-based approaches usually require large amounts of well-annotated data, which can be extremely expensive in the field of medical image analysis. Unlabeled data, on the other hand, is much easier to acquire. Semi-supervised learning and unsupervised domain adaptation both take the advantage of unlabeled data, and they are closely related to each other. In this paper, we propose uncertainty-aware multi-view co-training (UMCT), a unified framework that addresses these two tasks for volumetric medical image segmentation. Our framework is capable of efficiently utilizing unlabeled data for better performance. We firstly rotate and permute the 3D volumes into multiple views and train a 3D deep network on each view. We then apply co-training by enforcing multi-view consistency on unlabeled data, where an uncertainty estimation of each view is utilized to achieve accurate labeling. Experiments on the NIH pancreas segmentation dataset and a multi-organ segmentation dataset show state-of-the-art performance of the proposed framework on semi-supervised medical image segmentation. Under unsupervised domain adaptation settings, we validate the effectiveness of this work by adapting our multi-organ segmentation model to two pathological organs from the Medical Segmentation Decathlon Datasets. Additionally, we show that our UMCT-DA model can even effectively handle the challenging situation where labeled source data is inaccessible, demonstrating strong potentials for real-world applications.

[1]  Timo Aila,et al.  Temporal Ensembling for Semi-Supervised Learning , 2016, ICLR.

[2]  Shin Ishii,et al.  Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Yan Wang,et al.  Semi-Supervised Multi-Organ Segmentation via Deep Multi-Planar Co-Training , 2018, 1804.02586.

[4]  Hao Chen,et al.  Semi-supervised Skin Lesion Segmentation via Transformation Consistent Self-ensembling Model , 2018, BMVC.

[5]  Geoffrey French,et al.  Self-ensembling for visual domain adaptation , 2017, ICLR.

[6]  Konstantinos Kamnitsas,et al.  Unsupervised domain adaptation in brain lesion segmentation with adversarial networks , 2016, IPMI.

[7]  S. N. Merchant,et al.  MS-Net: Mixed-Supervision Fully-Convolutional Networks for Full-Resolution Segmentation , 2018, MICCAI.

[8]  Alan L. Yuille,et al.  Bridging the Gap Between 2D and 3D Organ Segmentation , 2018, MICCAI.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Iasonas Kokkinos,et al.  DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Martial Hebert,et al.  Semi-Supervised Self-Training of Object Detection Models , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[12]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[13]  Klaus H. Maier-Hein,et al.  nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation , 2018, Bildverarbeitung für die Medizin.

[14]  Tapani Raiko,et al.  Semi-supervised Learning with Ladder Networks , 2015, NIPS.

[15]  Ming-Hsuan Yang,et al.  Learning to Adapt Structured Output Space for Semantic Segmentation , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[16]  Trevor Darrell,et al.  Adapting Visual Category Models to New Domains , 2010, ECCV.

[17]  Miguel Á. Carreira-Perpiñán,et al.  Multiscale conditional random fields for image labeling , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[18]  Jonathon Shlens,et al.  Explaining and Harnessing Adversarial Examples , 2014, ICLR.

[19]  Trevor Darrell,et al.  FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation , 2016, ArXiv.

[20]  Andrew Blake,et al.  A framework for spatiotemporal control in the tracking of visual contours , 1993, International Journal of Computer Vision.

[21]  Ronald M. Summers,et al.  A large annotated medical image dataset for the development and evaluation of segmentation algorithms , 2019, ArXiv.

[22]  George Papandreou,et al.  Rethinking Atrous Convolution for Semantic Image Segmentation , 2017, ArXiv.

[23]  Ben Glocker,et al.  Semi-supervised Learning for Network-Based Cardiac MR Image Segmentation , 2017, MICCAI.

[24]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[25]  Harri Valpola,et al.  Weight-averaged consistency targets improve semi-supervised deep learning results , 2017, ArXiv.

[26]  Xiaofeng Liu,et al.  Confidence Regularized Self-Training , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[27]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[28]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[29]  Christian S. Perone,et al.  Unsupervised domain adaptation for medical imaging segmentation with self-ensembling , 2018, NeuroImage.

[30]  Yan Wang,et al.  Semi-supervised multi-organ segmentation via multi-planar co-training , 2018, ArXiv.

[31]  Chi-Wing Fu,et al.  H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes , 2018, IEEE Transactions on Medical Imaging.

[32]  Yaozong Gao,et al.  ASDNet: Attention Based Semi-supervised Deep Networks for Medical Image Segmentation , 2018, MICCAI.

[33]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[34]  Zhi-Hua Zhou,et al.  Semi-Supervised Regression with Co-Training , 2005, IJCAI.

[35]  Avrim Blum,et al.  The Bottleneck , 2021, Monopsony Capitalism.

[36]  Yarin Gal,et al.  Uncertainty in Deep Learning , 2016 .

[37]  Dean C. Barratt,et al.  Automatic Multi-Organ Segmentation on Abdominal CT With Dense V-Networks , 2018, IEEE Transactions on Medical Imaging.

[38]  Zhi-Hua Zhou,et al.  Tri-training: exploiting unlabeled data using three classifiers , 2005, IEEE Transactions on Knowledge and Data Engineering.

[39]  Yang Zou,et al.  Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training , 2018, ArXiv.

[40]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[41]  Dong Yang,et al.  3D Semi-Supervised Learning with Uncertainty-Aware Multi-View Co-Training , 2018, 2020 IEEE Winter Conference on Applications of Computer Vision (WACV).

[42]  Xiaogang Wang,et al.  Pyramid Scene Parsing Network , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Zhi-Hua Zhou,et al.  Tri-net for Semi-Supervised Deep Learning , 2018, IJCAI.

[44]  Ivor W. Tsang,et al.  Domain Adaptation via Transfer Component Analysis , 2009, IEEE Transactions on Neural Networks.

[45]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[46]  Taesung Park,et al.  CyCADA: Cycle-Consistent Adversarial Domain Adaptation , 2017, ICML.

[47]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[48]  Alan L. Yuille,et al.  Recurrent Saliency Transformation Network: Incorporating Multi-stage Visual Cues for Small Organ Segmentation , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[49]  Daguang Xu,et al.  3D Anisotropic Hybrid Network: Transferring Convolutional Features from 2D Images to 3D Anisotropic Volumes , 2017, MICCAI.

[50]  Tolga Tasdizen,et al.  Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning , 2016, NIPS.

[51]  Ronald M. Summers,et al.  Spatial aggregation of holistically‐nested convolutional neural networks for automated pancreas localization and segmentation☆ , 2017, Medical Image Anal..

[52]  Philip Bachman,et al.  Learning with Pseudo-Ensembles , 2014, NIPS.

[53]  Mikhail Belkin,et al.  Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples , 2006, J. Mach. Learn. Res..

[54]  Hao Chen,et al.  Unsupervised Cross-Modality Domain Adaptation of ConvNets for Biomedical Image Segmentations with Adversarial Loss , 2018, IJCAI.

[55]  Yuanqing Li,et al.  A self-training semi-supervised SVM algorithm and its application in an EEG-based brain computer interface speller system , 2008, Pattern Recognit. Lett..

[56]  Abhishek Kumar,et al.  Semi-supervised Learning with GANs: Manifold Invariance with Improved Inference , 2017, NIPS.

[57]  Kate Saenko,et al.  Subspace Distribution Alignment for Unsupervised Domain Adaptation , 2015, BMVC.

[58]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[59]  Josien P. W. Pluim,et al.  Not‐so‐supervised: A survey of semi‐supervised, multi‐instance, and transfer learning in medical image analysis , 2018, Medical Image Anal..

[60]  Yuan Shi,et al.  Geodesic flow kernel for unsupervised domain adaptation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[61]  Bo Wang,et al.  Deep Co-Training for Semi-Supervised Image Recognition , 2018, ECCV.

[62]  Joseph O. Deasy,et al.  Tumor-Aware, Adversarial Domain Adaptation from CT to MRI for Lung Cancer Segmentation , 2018, MICCAI.

[63]  Yan Wang,et al.  A Fixed-Point Model for Pancreas Segmentation in Abdominal CT Scans , 2016, MICCAI.

[64]  Hao Chen,et al.  Volumetric ConvNets with Mixed Residual Connections for Automated Prostate Segmentation from 3D MR Images , 2017, AAAI.

[65]  Hervé Delingette,et al.  Deep learning with mixed supervision for brain tumor segmentation , 2018, Journal of medical imaging.

[66]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[67]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[68]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[69]  Ronald M. Summers,et al.  DeepOrgan: Multi-level Deep Convolutional Networks for Automated Pancreas Segmentation , 2015, MICCAI.

[70]  Stefano Ermon,et al.  A DIRT-T Approach to Unsupervised Domain Adaptation , 2018, ICLR.

[71]  Michael Kampffmeyer,et al.  Unsupervised Domain Adaptation for Automatic Estimation of Cardiothoracic Ratio , 2018, MICCAI.

[72]  Fan Yang,et al.  Good Semi-supervised Learning That Requires a Bad GAN , 2017, NIPS.

[73]  John Blitzer,et al.  Co-Training for Domain Adaptation , 2011, NIPS.

[74]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.