Boundary-enhanced semi-supervised retinal layer segmentation in optical coherence tomography images using fewer labels. (April 2023)

• Record Type: Journal Article
• Authors: Lu, Ye; Shen, Yutian; Xing, Xiaohan; Ye, Chengwei; Meng, Max Q.-H.
• Is Part Of: Computerized medical imaging and graphics. Volume 105(2023)
• Publication Date: 2023-04
• Subjects: Optical coherence tomography -- Retinal layer segmentation -- Semi-supervised learning -- Distance transform map; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2023.102199
• Languages: English
• ISSNs: 0895-6111

Abstract

Abstract: Automatic segmentation of multiple layers in retinal optical coherence tomography (OCT) images is crucial for eye disease diagnosis and treatment. Despite the success of deep learning algorithms, it still remains a challenge due to the blurry layer boundaries and lack of adequate pixel-wise annotations. To tackle these issues, we propose a Boundary-Enhanced Semi-supervised Network (BE-SemiNet) that exploits an auxiliary distance map regression task to improve retinal layer segmentation with scarce labeled data and abundant unlabeled data. Specifically, a novel Unilaterally Truncated Distance Map (UTDM) is firstly introduced to alleviate the class imbalance problem and enhance the layer boundary learning in the regression task. Then for the pixel-wise segmentation and UTDM regression branches, we impose task-level and data-level consistency regularization on unlabeled data to enrich the diversity of unsupervised information and improve the regularization effects. Pseudo supervision is incorporated in consistency regularization to bridge the task prediction spaces for consistency and expand training labeled data. Experiments on two public retinal OCT datasets show that our method can greatly improve the supervised baseline performance with only 5 annotations and outperform the state-of-the-art methods. Since it is difficult and labor-expensive to obtain adequate pixel-wise annotations in practice, our method has a promising application future in clinical retinal OCT image analysis. Highlights: A novel distance map is devised for regression to enhance layer boundary learning. A dual-task network is developed to enable boundary-enhanced segmentation. Semi-supervised learning is imposed via task and data level consistency supervision. Pseudo supervision is employed for consistency regularization to bridge task space.