Segmentation Only Uses Sparse Annotations: Unified Weakly and Semi-Supervised Learning in Medical Images

• Published in: Medical image analysis 2022-08, Vol.80, p.102515-102515, Article 102515
• Main Authors: Gao, Feng, Hu, Minhao, Zhong, Min-Er, Feng, Shixiang, Tian, Xuwei, Meng, Xiaochun, Ni-jia-ti, Ma-yi-di-li, Huang, Zeping, Lv, Minyi, Song, Tao, Zhang, Xiaofan, Zou, Xiaoguang, Wu, Xiaojian
• Format: Article
• Language: English
• Subjects: Medical image; Semantic segmentation; Semi-supervised learning; Weakly supervised learning
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2022.102515

•We designed a image segmentation framework with weakly- semi- supervision setting.•We proposed a consistency loss, which amplifies the error of boundary and outliers.•We achieved a comparable performance with fully supervised models on two datasets. [Display omitted] Since segmentation labeling is usually time-consuming and annotating medical images requires professional expertise, it is laborious to obtain a large-scale, high-quality annotated segmentation dataset. We propose a novel weakly- and semi-supervised framework named SOUSA (Segmentation Only Uses Sparse Annotations), aiming at learning from a small set of sparse annotated data and a large amount of unlabeled data. The proposed framework contains a teacher model and a student model. The student model is weakly supervised by scribbles and a Geodesic distance map derived from scribbles. Meanwhile, a large amount of unlabeled data with various perturbations are fed to student and teacher models. The consistency of their output predictions is imposed by Mean Square Error (MSE) loss and a carefully designed Multi-angle Projection Reconstruction (MPR) loss. Extensive experiments are conducted to demonstrate the robustness and generalization ability of our proposed method. Results show that our method outperforms weakly- and semi-supervised state-of-the-art methods on multiple datasets. Furthermore, our method achieves a competitive performance with some fully supervised methods with dense annotation when the size of the dataset is limited.