JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]


Journal of Information Science and Engineering, Vol. 40 No. 5, pp. 1071-1092


Semi-Supervised Learning Using Co-Generative Adversarial Network (Co-GAN) for Medical Image Segmentation


GUO-QIN LI1,2, NURSURIATI JAMIL2 AND RASEEDA HAMZAH3,+
1Taiyuan Institute of Technology
Taiyuan, 030008 P.R. China

2College of Computing, Informatics and Media
Universiti Teknologi MARA (UiTM)
Selangor, 40450 Malaysia

3College of Computing, Informatics and Media
Universiti Teknologi MARA (UiTM), Melaka Branch
Merlimau, Melaka, 77300 Malaysia
E-mail: raseeda@uitm.edu.my
+


Medical image analysis has experienced different stages of development, especially with the emergence of deep learning. However, acquiring large-scale, high-quality labeled data to train a deep learning model takes time and effort. This paper proposes a semi-supervised learning method for medical image segmentation using limited labeled data and large-scale unlabeled data. Inspired by the classic Generative Adversarial Network (GAN) and co-training strategy, we proposed a new Co-GAN framework to implement medical image segmentation. The proposed Co-GAN comprises two generators and one discrimi-nator, in which two generators can provide mutual segmentation information to each other. Through adversarial training between generators and discriminators, Co-GAN achieved higher segmentation accuracy. The dataset used was the hippocampus in Medical Segmen-tation Decathlon (MSD). There were four training data settings: 25 labeled slices/3,374 unlabeled slices; 50 labeled slices/3,349 unlabeled slices; 100 labeled slices/3,299 unla-beled slices; and 200 labeled slices/3,199 unlabeled slices. Three experiments were con-ducted for each data set: fully supervised learning based on a generator network using only labeled data (F-Generator), semi-supervised learning based on GAN (Semi-GAN), and semi-supervised learning based on Co-GAN. The experiments showed that Co-GAN im-proved the segmentation accuracy by (1.9%, 2.6%, 1.1%, and 0.1%) compared to F-Gen-erator and (2.2%, 0.8%, 0.5%, 0.7%) to Semi-GAN.


Keywords: semi-supervised learning, GAN, co-training, Co-GAN, medical image segmentation

  Retrieve PDF document (JISE_202405_10.pdf)