Data Augmentation Using Learned Transformations for One-Shot Medical Image Segmentation
By Amy Zhao, Guha Balakrishnan, Frédo Durand, John V. Guttag, Adrian V. Dalca. 2019
448 citations recorded by the source corpus.
Image segmentation is an important task in many medical applications. Methods based on convolutional neural networks attain state-of-the-art accuracy; however, they typically rely on supervised training with large labeled datasets. Labeling medical images requires significant expertise and time, and typical hand-tuned approaches for data augmentation fail to capture the complex variations in such images. We present an automated data augmentation method for synthesizing labeled medical images. We demonstrate our method on the task of segmenting magnetic resonance imaging (MRI) brain scans. Our method requires only a single segmented scan, and leverages other unlabeled scans in a semi-supervised approach. We learn a model of transformations from the images, and use the model along with the labeled example to synthesize additional labeled examples. Each transformation is comprised of a spatial deformation field and an intensity change, enabling the synthesis of complex effects such as variations in anatomy and image acquisition procedures. We show that training a supervised segmenter with these new examples provides significant improvements over state-of-the-art methods for one-shot biomedical image segmentation.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Medical Imaging, Meta & Continual Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Katja Hofmann
- Luisa M. Zintgraf
- John F. Bronskill
- Daniela Massiceti
- Ender Konukoglu
- Cecily Morrison
- Tao Lei
- Chen Qin
In Computer Vision and Pattern Recognition
- Ender Konukoglu
- Daniel Rueckert
- Wenjia Bai
- Cecily Morrison
- Tao Lei
- Chen Qin
- Bin Zheng
- Jyoti Prakash Sahoo
In Medical Imaging
In Meta & Continual Learning
- Katja Hofmann
- Luisa M. Zintgraf
- Daniela Massiceti
- Jyoti Prakash Sahoo
- Yisheng Song
- Ting-Yuan Wang
- Subrota Kumar Mondal
- Puyu Cai
Citations in the corpus
- ORBIT: A Real-World Few-Shot Dataset for Teachable Object Recognition (2021, background): “[53] Amy Zhao, Guha Balakrishnan, Fredo Durand, John V Gut-”
- Difficulty-aware Meta-Learning for Rare Disease Diagnosis (2019, unknown): “[15] and Mondal et al .”
- Recent advances and clinical applications of deep learning in medical image analysis (2021, background): “For medical image segmentation, to alleviate the need for a large amount of annotated training data, reserachers have adopted generative models for image synthesis to increase the number of training examples (Zhang et al., 2018b; Zhao et al., 2019a).”
- Deep Learning for Cardiac Image Segmentation: A Review (2019, methodology): “In particular, the generative model-based approach has been proven to be effective for one-shot brain segmentation (Zhao et al., 2019) and few-shot cardiac MR image segmentation (Chaitanya et al.”
- Contrastive learning of global and local features for medical image segmentation with limited annotations (2020, background): “works also leveraged unlabeled data for augmentation either through image registration [64] or, in more general way, optimizing the augmentation procedures for a given task [9].”
- SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining (2021, unknown): “Therefore, there is a clear need for a fast, accurate, and reproducible automated method, for segmentation of brain scans of any contrast and resolution, and that can adapt to a wide range of populations.”
- Unsupervised Intra-Domain Adaptation for Semantic Segmentation Through Self-Supervision (2020, background): “Recently, convolutional neural network-based segmentation models [13, 33] have achieved remarkable progresses, leading to various applications in computer vision systems, such as autonomous driving [14, 31], robotics [15, 23], and disease diagnosis [35, 32].”
- Medical Image Segmentation Using Deep Learning: A Survey (2020, background): “Similarly, some other semi-supervised frameworks [97-100] are also proposed to optimize medical image segmentation.”
- A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, background): “Global Class Representations [77] ICCV’19 Data Expansion mini-ImageNet* ✓ AutoAugment [52] CVPR’19 Data Expansion CIFAR10* EDA [53] EMNLP IJCNLP’19 Data Expansion SST-2 LaSO [70] CVPR’19 Feature Augmentation Mlearned ✓ IDeMe-Net [54] CVPR’19 Data Expansion mini-ImageNet* Learned Transformations[59] ”
- Data Augmentation for Graph Neural Networks (2020, methodology): “…data without additional ground-truth labels, and have seen widespread adoption in fields such as computer vision (CV) (DeVries and Taylor 2017; Cubuk et al. 2019; Zhao et al. 2019; Ho et al. 2019), and natural language processing (NLP) (Fadaee, Bisazza, and Monz 2017; Şahin and Steedman 2019).”