Abstract
Convolutional Neural Networks (CNNs) can achieve excellent computer-assisted diagnosis performance, relying on sufficient annotated training data. Unfortunately, most medical imaging datasets, often collected from various scanners, are small and fragmented. In this context, as a Data Augmentation (DA) technique, Generative Adversarial Networks (GANs) can synthesize realistic/diverse additional training images to fill the data lack in the real image distribution; researchers have improved classification by augmenting images with noise-to-image (e.g., random noise samples to diverse pathological images) or image-to-image GANs (e.g., a benign image to a malignant one). Yet, no research has reported results combining (i) noise-to-image GANs and image-to-image GANs or (ii) GANs and other deep generative models, for further performance boost. Therefore, to maximize the DA effect with the GAN combinations, we propose a two-step GAN-based DA that generates and refines brain MR images with/without tumors separately: (i) Progressive Growing of GANs (PGGANs), multi-stage noise-to-image GAN for high-resolution image generation, first generates realistic/diverse 256 x 256 images--even a physician cannot accurately distinguish them from real ones via Visual Turing Test; (ii) UNsupervised Image-to-image Translation or SimGAN, image-to-image GAN combining GANs/Variational AutoEncoders or using a GAN loss for DA, further refines the texture/shape of the PGGAN-generated images similarly to the real ones. We thoroughly investigate CNN-based tumor classification results, also considering the influence of pre-training on ImageNet and discarding weird-looking GAN-generated images. The results show that, when combined with classic DA, our two-step GAN-based DA can significantly outperform the classic DA alone, in tumor detection (i.e., boosting sensitivity from 93.63% to 97.53%) and also in other tasks.
Abstract (translated)
卷积神经网络（CNN）依靠足够的标注训练数据，可以获得良好的计算机辅助诊断性能。不幸的是，大多数医学成像数据集，通常是从不同的扫描仪收集的，都很小而且支离破碎。在此背景下，生成对抗网络（gans）作为一种数据增强（da）技术，可以合成真实/多样的额外训练图像，以填补真实图像分布中的数据不足；研究人员通过将图像与噪声进行图像增强来改进分类（例如，随机噪声样本与多样的病理图像）。）或图像到图像的gan（例如，良性图像到恶性图像）。然而，还没有研究报告将（i）噪声与图像Gans和图像与图像Gans相结合的结果，或（ii）Gans和其他深层次生成模型，以进一步提高性能。因此，为了最大限度地利用氮化镓的结合，我们提出了一种基于氮化镓的两步DA，它分别生成和优化有/无肿瘤的大脑磁共振图像：（i）Gans（pggans）的渐进式生长，高分辨率图像生成的多级噪声到图像Gan，首先生成逼真/多样的256 x 256图像，甚至是一个物理图像。CIAN无法通过视觉图灵测试准确地将其与真实图像区分开来；（ii）无监督图像到图像转换或simgan、图像到图像gan结合gans/变分自动编码器或使用gan丢失进行da，进一步优化pggan生成的图像的纹理/形状，与真实图像相似。我们深入研究了基于CNN的肿瘤分类结果，同时考虑了预训练对图像网络的影响，丢弃了GaN生成的奇怪图像。结果表明，在肿瘤检测（即将敏感性从93.63%提高到97.53%）以及其他任务中，我们的两步Gan-based DA与经典DA结合后，可以显著优于经典DA。
URL
https://arxiv.org/abs/1905.13456
