Edge-Guided Output Adaptor: Highly Efficient Adaptation Module for Cross-Vendor Medical Image Segmentation

Supervised convolutional neural networks (CNNs) have demonstrated state-of-art performance in medical image segmentation tasks. However, the performance of a well-trained CNN on an independent dataset (e.g., different vendors, sequences) relies strongly on the distribution similarity, and may drop unexpectedly in case of distribution shift. To obtain a large amount of annotation from each new dataset for re-training the CNN is expensive and impractical. Adaptation algorithms to improve the CNN generalizability from source domain to target domain has significant practical value. In this work, we propose a highly efficient end-to-end domain adaptation approach, with left ventricle segmentation from cine MRI sequences as an example. We propose to perform domain adaptation in the output space where different domains share the strongest similarities. The core of this algorithm is a flexible and light output adaption module based on adversarial learning. Moreover, Canny edge detector is introduced to enhance model's attention to edges during adversarial learning. Comparative experiments were carried out using images from three major MR vendors (Philips, Siemens, and GE) as three domains. Our results demonstrated that the proposed method substantially improved the generalization of the trained CNN model from one vendor to other vendors without any additional annotation. Moreover, the ablation study proved that introducing Canny edge detector further refined the edge detection in segmentation. The proposed adaption is generic can be extended to other medical image segmentation problems.

[1]  Jefersson Alex dos Santos,et al.  Conditional Domain Adaptation GANs for Biomedical Image Segmentation , 2019, ArXiv.

[2]  Xin Qi,et al.  Adversarial Domain Adaptation for Classification of Prostate Histopathology Whole-Slide Images , 2018, MICCAI.

[3]  Tolga Tasdizen,et al.  Domain adaptation for biomedical image segmentation using adversarial training , 2018, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018).

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

[5]  Trevor Darrell,et al.  Adversarial Discriminative Domain Adaptation , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Dumitru Erhan,et al.  Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[8]  Pedro H. O. Pinheiro,et al.  Unsupervised Domain Adaptation with Similarity Learning , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[9]  Shuiwang Ji,et al.  Residual Deconvolutional Networks for Brain Electron Microscopy Image Segmentation , 2017, IEEE Transactions on Medical Imaging.

[10]  Linda G. Shapiro,et al.  Multi-Instance Multi-Label Learning for Multi-Class Classification of Whole Slide Breast Histopathology Images , 2018, IEEE Transactions on Medical Imaging.

[11]  Nico Karssemeijer,et al.  Transfer Learning for Domain Adaptation in MRI: Application in Brain Lesion Segmentation , 2017, MICCAI.

[12]  Ben Glocker,et al.  Automated cardiovascular magnetic resonance image analysis with fully convolutional networks , 2017, Journal of Cardiovascular Magnetic Resonance.

[13]  W. Marsden I and J , 2012 .

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

[15]  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).

[16]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  S. Plein,et al.  Deep Learning-based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study. , 2019, Radiology.

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

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

[20]  Hao Chen,et al.  3D deeply supervised network for automated segmentation of volumetric medical images , 2017, Medical Image Anal..

[21]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[22]  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.

[23]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

[24]  Lorenzo Bruzzone,et al.  Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances , 2016, IEEE Geoscience and Remote Sensing Magazine.

[25]  Philip J. Morrow,et al.  Fully automated breast boundary and pectoral muscle segmentation in mammograms , 2017, Artif. Intell. Medicine.

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

[27]  Yue Zhang,et al.  Task Driven Generative Modeling for Unsupervised Domain Adaptation: Application to X-ray Image Segmentation , 2018, MICCAI.

[28]  Kurt Keutzer,et al.  A Novel Domain Adaptation Framework for Medical Image Segmentation , 2018, BrainLes@MICCAI.

[29]  Leyuan Fang,et al.  Automatic Classification of Retinal Optical Coherence Tomography Images With Layer Guided Convolutional Neural Network , 2019, IEEE Signal Processing Letters.

[30]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.