AIDE: Annotation-efficient deep learning for automatic medical image segmentation

Accurate image segmentation is crucial for medical imaging applications. The prevailing deep learning approaches typically rely on very large training datasets with high-quality manual annotations, which are often not available in medical imaging. We introduce Annotation-effIcient Deep lEarning (AIDE) to handle imperfect datasets with an elaborately designed cross-model self-correcting mechanism. AIDE improves the segmentation Dice scores of conventional deep learning models on open datasets possessing scarce or noisy annotations by up to 30%. For three clinical datasets containing 11,852 breast images of 872 patients from three medical centers, AIDE consistently produces segmentation maps comparable to those generated by the fully supervised counterparts as well as the manual annotations of independent radiologists by utilizing only 10% training annotations. Such a 10-fold improvement of efficiency in utilizing experts' labels has the potential to promote a wide range of biomedical applications.

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

[2]  Haidong Zhu,et al.  Pick-and-Learn: Automatic Quality Evaluation for Noisy-Labeled Image Segmentation , 2019, MICCAI.

[3]  Z. Fayad,et al.  Artificial intelligence–enabled rapid diagnosis of patients with COVID-19 , 2020, Nature Medicine.

[4]  Hui Sun,et al.  Learning Cross-Modal Deep Representations for Multi-Modal MR Image Segmentation , 2019, MICCAI.

[5]  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).

[6]  Dong Yang,et al.  Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation , 2020, Medical Image Anal..

[7]  Nima Tajbakhsh,et al.  Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation , 2019, Medical Image Anal..

[8]  Pengfei Chen,et al.  Understanding and Utilizing Deep Neural Networks Trained with Noisy Labels , 2019, ICML.

[9]  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.

[10]  Xiaohui Xie,et al.  Clinically applicable deep learning framework for organs at risk delineation in CT images , 2019, Nature Machine Intelligence.

[11]  Dong-Hyun Lee,et al.  Pseudo-Label : The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks , 2013 .

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

[13]  Sarah Webb.  Deep learning for biology , 2018, Nature.

[14]  Yong Wang,et al.  Augmenting vascular disease diagnosis by vasculature-aware unsupervised learning , 2020, Nature Machine Intelligence.

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

[16]  Xingrui Yu,et al.  How does Disagreement Help Generalization against Label Corruption? , 2019, ICML.

[17]  Christopher Churas,et al.  CDeep3M—Plug-and-Play cloud-based deep learning for image segmentation , 2018, Nature Methods.

[18]  Yike Guo,et al.  A population-based phenome-wide association study of cardiac and aortic structure and function , 2020, Nature Medicine.

[19]  Daniel Cremers,et al.  FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-Based CNN Architecture , 2016, ACCV.

[20]  Naciye Sinem Gezer,et al.  Comparison of semi-automatic and deep learning-based automatic methods for liver segmentation in living liver transplant donors. , 2020, Diagnostic and interventional radiology.

[21]  Samy Bengio,et al.  Understanding deep learning requires rethinking generalization , 2016, ICLR.

[22]  Xingrui Yu,et al.  Co-teaching: Robust training of deep neural networks with extremely noisy labels , 2018, NeurIPS.

[23]  Hao Chen,et al.  Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation , 2020, IEEE Transactions on Medical Imaging.

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

[25]  Andre Esteva,et al.  A guide to deep learning in healthcare , 2019, Nature Medicine.

[26]  Wu Xiao,et al.  LVC-Net: Medical Image Segmentation with Noisy Label Based on Local Visual Cues. , 2019 .

[27]  Geraint Rees,et al.  Clinically applicable deep learning for diagnosis and referral in retinal disease , 2018, Nature Medicine.

[28]  Dinggang Shen,et al.  Interleaved 3D-CNNs for joint segmentation of small-volume structures in head and neck CT images. , 2018, Medical physics.

[29]  Joel H. Saltz,et al.  Pancreatic Cancer Detection in Whole Slide Images Using Noisy Label Annotations , 2019, MICCAI.

[30]  Thomas Brox,et al.  U-Net: deep learning for cell counting, detection, and morphometry , 2018, Nature Methods.

[31]  Simon K. Warfield,et al.  Deep learning with noisy labels: exploring techniques and remedies in medical image analysis , 2020, Medical Image Anal..

[32]  Florian Jung,et al.  Evaluation of prostate segmentation algorithms for MRI: The PROMISE12 challenge , 2014, Medical Image Anal..

[33]  Christoph Meinel,et al.  Deep Learning for Medical Image Analysis , 2018, Journal of Pathology Informatics.

[34]  Yoshua Bengio,et al.  A Closer Look at Memorization in Deep Networks , 2017, ICML.

[35]  Andreas Nürnberger,et al.  CHAOS Challenge - Combined (CT-MR) Healthy Abdominal Organ Segmentation , 2020, Medical Image Anal..

[36]  Daniel Rueckert,et al.  Self-Supervised Learning for Cardiac MR Image Segmentation by Anatomical Position Prediction , 2019, MICCAI.

[37]  Li Fei-Fei,et al.  MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels , 2017, ICML.

[38]  Ganapathy Krishnamurthi,et al.  Fully convolutional multi‐scale residual DenseNets for cardiac segmentation and automated cardiac diagnosis using ensemble of classifiers , 2018, Medical Image Anal..

[39]  R. B. Bernstein,et al.  Achievements and Challenges , 2011 .

[40]  Kaiming He,et al.  Data Distillation: Towards Omni-Supervised Learning , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[41]  Xiangjian He,et al.  Deep Learning Techniques for Medical Image Segmentation: Achievements and Challenges , 2019, Journal of Digital Imaging.

[42]  John E. Tomaszewski,et al.  An integrated iterative annotation technique for easing neural network training in medical image analysis , 2019, Nat. Mach. Intell..

[43]  Nima Tajbakhsh,et al.  Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis , 2019, MICCAI.

[44]  Jing Yuan,et al.  HyperDense-Net: A Hyper-Densely Connected CNN for Multi-Modal Image Segmentation , 2018, IEEE Transactions on Medical Imaging.

[45]  Ronald M. Summers,et al.  Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique , 2016 .

[46]  Lin Yang,et al.  Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

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

[48]  Konstantinos Kamnitsas,et al.  Overfitting of neural nets under class imbalance: Analysis and improvements for segmentation , 2019, MICCAI.