GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification
GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification
复制标题
DOI：
10.1016/j.neucom.2018.09.013
复制
发表时间：
2018-12-10
期刊：
影响因子：
6
通讯作者：
Greenspan, Hayit
中科院分区：
文献类型：
--
作者：
Frid-Adar, Maayan;Diamant, Idit;Greenspan, Hayit
深度学习方法，特别是卷积神经网络（cnn），主要通过使用大规模注释数据集，在广泛的计算机视觉任务中取得了巨大的突破。然而，在医学领域获得这样的数据集仍然是一个挑战。在本文中，我们提出了使用最近提出的深度学习生成对抗网络（GANs）生成合成医学图像的方法。此外，我们还证明了生成的医学图像可以用于合成数据增强，并提高了CNN在医学图像分类方面的性能。我们的新方法在182个肝脏病变（53个囊肿，64个转移瘤和65个血管瘤）的计算机断层扫描（CT）图像的有限数据集上得到了验证。我们首先利用GAN架构来合成高质量的肝脏病变roi。然后，我们提出了一种新的基于CNN的肝脏病变分类方案。最后，我们使用经典数据增强和我们的合成数据增强对CNN进行训练，并比较性能。此外，我们利用可视化和专家评估来探索我们合成示例的质量。仅使用经典数据增强的分类性能为78.6%的敏感性和88.4%的特异性。通过添加合成数据，结果敏感性提高到85.7%，特异性提高到92.4%。我们相信这种合成数据增强的方法可以推广到其他医学分类应用中，从而支持放射科医生提高诊断的努力。(C) 2018 Elsevier B.V.版权所有
Deep learning methods, and in particular convolutional neural networks (CNNs), have led to an enormous breakthrough in a wide range of computer vision tasks, primarily by using large-scale annotated datasets. However, obtaining such datasets in the medical domain remains a challenge. In this paper, we present methods for generating synthetic medical images using recently presented deep learning Generative Adversarial Networks (GANs). Furthermore, we show that generated medical images can be used for synthetic data augmentation, and improve the performance of CNN for medical image classification. Our novel method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). We first exploit GAN architectures for synthesizing high quality liver lesion ROIs. Then we present a novel scheme for liver lesion classification using CNN. Finally, we train the CNN using classic data augmentation and our synthetic data augmentation and compare performance. In addition, we explore the quality of our synthesized examples using visualization and expert assessment. The classification performance using only classic data augmentation yielded 78.6% sensitivity and 88.4% specificity. By adding the synthetic data augmentation the results increased to 85.7% sensitivity and 92.4% specificity. We believe that this approach to synthetic data augmentation can generalize to other medical classification applications and thus support radiologists' efforts to improve diagnosis. (C) 2018 Elsevier B.V. All rights reserved.