MTL-ABS3Net: Atlas-Based Semi-Supervised Organ Segmentation Network With Multi-Task Learning for Medical Images

• Published in: IEEE journal of biomedical and health informatics 2022-08, Vol.26 (8), p.3988-3998
• Main Authors: Huang, Huimin, Chen, Qingqing, Lin, Lanfen, Cai, Ming, Zhang, Qiaowei, Iwamoto, Yutaro, Han, Xianhua, Furukawa, Akira, Kanasaki, Shuzo, Chen, Yen-Wei, Tong, Ruofeng, Hu, Hongjie
• Format: Article
• Language: English
• Subjects: Anatomical Priors; Biomedical imaging; Image analysis; Image processing; Image reconstruction; Image segmentation; Labels; Learning; Localization; Medical imaging; Multi-task Learning; Multitasking; Organ Segmentation; Probabilistic logic; Reconstruction; Self-training; Semi-supervised learning; Semi-supervised Learning (SSL); Task analysis; Training
• ISSN: 2168-2194; 2168-2208
• DOI: 10.1109/JBHI.2022.3153406

Organ segmentation is one of the most important step for various medical image analysis tasks. Recently, semi-supervised learning (SSL) has attracted much attentions by reducing labeling cost. However, most of the existing SSLs neglected the prior shape and position information specialized in the medical images, leading to unsatisfactory localization and non-smooth of objects. In this paper, we propose a novel atlas-based semi-supervised segmentation network with multi-task learning for medical organs, named MTL-ABS 3 Net, which incorporates the anatomical priors and makes full use of unlabeled data in a self-training and multi-task learning manner. The MTL-ABS 3 Net consists of two components: an Atlas-Based Semi-Supervised Segmentation Network (ABS 3 Net) and Reconstruction-Assisted Module (RAM). Specifically, the ABS 3 Net improves the existing SSLs by utilizing atlas prior, which generates credible pseudo labels in a self-training manner; while the RAM further assists the segmentation network by capturing the anatomical structures from the original images in a multi-task learning manner. Better reconstruction quality is achieved by using MS-SSIM loss function, which further improves the segmentation accuracy. Experimental results from the liver and spleen datasets demonstrated that the performance of our method was significantly improved compared to existing state-of-the-art methods.