Few-shot medical image segmentation using a global correlation network with discriminative embedding. (January 2022)

• Record Type: Journal Article
• Authors: Sun, Liyan; Li, Chenxin; Ding, Xinghao; Huang, Yue; Chen, Zhong; Wang, Guisheng; Yu, Yizhou; Paisley, John
• Is Part Of: Computers in biology and medicine. Volume 140(2022)
• Publication Date: 2022-01
• Subjects: Few-shot learning -- Medical image segmentation -- Cross correlation -- Deep embedding; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.105067
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Despite impressive developments in deep convolutional neural networks for medical imaging, the paradigm of supervised learning requires numerous annotations in training to avoid overfitting. In clinical cases, massive semantic annotations are difficult to acquire where biomedical expert knowledge is required. Moreover, it is common when only a few annotated classes are available. In this study, we proposed a new approach to few-shot medical image segmentation, which enables a segmentation model to quickly generalize to an unseen class with few training images. We constructed a few-shot image segmentation mechanism using a deep convolutional network trained episodically. Motivated by the spatial consistency and regularity in medical images, we developed an efficient global correlation module to model the correlation between a support and query image and incorporate it into the deep network. We enhanced the discrimination ability of the deep embedding scheme to encourage clustering of feature domains belonging to the same class while keeping feature domains of different organs far apart. We experimented using anatomical abdomen images from both CT and MRI modalities. Graphical abstract: Image 1 Highlights: We proposed an efficient global correlation module to capture the correspondence between the support and query image pair. We enhanced feature learning by discriminative embedding for intra-compactness and inter-class separability. We developed a modified episodic training to adapt to the discriminative embedding. The proposed model achieved promising results in few-shot abdominal organ segmentation.