Generation of 3D Brain MRI Using Auto-Encoding Generative Adversarial Networks

被引:68
|
作者
Kwon, Gihyun [1 ]
Han, Chihye [1 ]
Kim, Dae-shik [1 ]
机构
[1] Korea Adv Inst Sci & Technol, Daejeon, South Korea
基金
新加坡国家研究基金会;
关键词
Generative Adversarial Networks; MRI; Data augmentation; 3D; Image synthesis;
D O I
10.1007/978-3-030-32248-9_14
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
As deep learning is showing unprecedented success in medical image analysis tasks, the lack of sufficient medical data is emerging as a critical problem. While recent attempts to solve the limited data problem using Generative Adversarial Networks (GAN) have been successful in generating realistic images with diversity, most of them are based on image-to-image translation and thus require extensive datasets from different domains. Here, we propose a novel model that can successfully generate 3D brain MRI data from random vectors by learning the data distribution. Our 3D GAN model solves both image blurriness and mode collapse problems by leveraging alpha-GAN that combines the advantages of Variational Auto-Encoder (VAE) and GAN with an additional code discriminator network. We also use the Wasserstein GAN with Gradient Penalty (WGAN-GP) loss to lower the training instability. To demonstrate the effectiveness of our model, we generate new images of normal brain MRI and show that our model outperforms baseline models in both quantitative and qualitative measurements. We also train the model to synthesize brain disorder MRI data to demonstrate the wide applicability of our model. Our results suggest that the proposed model can successfully generate various types and modalities of 3D whole brain volumes from a small set of training data.
引用
收藏
页码:118 / 126
页数:9
相关论文
共 50 条
  • [31] 3D Point Cloud Shape Generation with Collaborative Learning of Generative Adversarial Network and Auto-Encoder
    Yang, Dong
    Wang, Jingyuan
    Yang, Xi
    REMOTE SENSING, 2024, 16 (10)
  • [32] Shape Inpainting using 3D Generative Adversarial Network and Recurrent Convolutional Networks
    Wang, Weiyue
    Huang, Qiangui
    You, Suya
    Yang, Chao
    Neumann, Ulrich
    2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, : 2317 - 2325
  • [33] 3D Design Using Generative Adversarial Networks and Physics-Based Validation
    Shu, Dule
    Cunningham, James
    Stump, Gary
    Miller, Simon W.
    Yukish, Michael A.
    Simpson, Timothy W.
    Tucker, Conrad S.
    JOURNAL OF MECHANICAL DESIGN, 2020, 142 (07)
  • [34] 3D image augmentation using Neural Style Transfer and Generative Adversarial Networks
    Cuong Do
    APPLICATIONS OF DIGITAL IMAGE PROCESSING XLIII, 2020, 11510
  • [35] Unregularized Auto-Encoder with Generative Adversarial Networks for Image Generation
    Wang, Jiayu
    Zhou, Wengang
    Tang, Jinhui
    Fu, Zhongqian
    Tian, Qi
    Li, Houqiang
    PROCEEDINGS OF THE 2018 ACM MULTIMEDIA CONFERENCE (MM'18), 2018, : 709 - 717
  • [36] Subsampled brain MRI reconstruction by generative adversarial neural networks
    Shaul, Roy
    David, Itamar
    Shitrit, Ohad
    Raviv, Tammy Riklin
    MEDICAL IMAGE ANALYSIS, 2020, 65
  • [37] The role of generative adversarial networks in brain MRI: a scoping review
    Hazrat Ali
    Md. Rafiul Biswas
    Farida Mohsen
    Uzair Shah
    Asma Alamgir
    Osama Mousa
    Zubair Shah
    Insights into Imaging, 13
  • [38] Fusion of Complementary 2D and 3D Mesostructural Datasets Using Generative Adversarial Networks
    Dahari, Amir
    Kench, Steve
    Squires, Isaac
    Cooper, Samuel J.
    ADVANCED ENERGY MATERIALS, 2023, 13 (02)
  • [39] 3D visualization model construction based on generative adversarial networks
    Liu, Xiaojuan
    Zhou, Shangbo
    Wu, Sheng
    Tan, Duo
    Yao, Rui
    PEERJ COMPUTER SCIENCE, 2022, 8
  • [40] Frequency-Domain Acceleration for 3D Generative Adversarial Networks
    Jiang Z.
    Liu B.
    Tang Y.
    Wu J.
    Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics, 2023, 35 (06): : 953 - 960