Graduate School of Health Sciences, Fujita Health University, Aichi, Japan.
Graduate School of Informatics, Nagoya University, Aichi, Japan.
Sci Rep. 2022 Jul 27;12(1):12867. doi: 10.1038/s41598-022-16861-5.
Artificial intelligence (AI) applications in medical imaging continue facing the difficulty in collecting and using large datasets. One method proposed for solving this problem is data augmentation using fictitious images generated by generative adversarial networks (GANs). However, applying a GAN as a data augmentation technique has not been explored, owing to the quality and diversity of the generated images. To promote such applications by generating diverse images, this study aims to generate free-form lesion images from tumor sketches using a pix2pix-based model, which is an image-to-image translation model derived from GAN. As pix2pix, which assumes one-to-one image generation, is unsuitable for data augmentation, we propose StylePix2pix, which is independently improved to allow one-to-many image generation. The proposed model introduces a mapping network and style blocks from StyleGAN. Image generation results based on 20 tumor sketches created by a physician demonstrated that the proposed method can reproduce tumors with complex shapes. Additionally, the one-to-many image generation of StylePix2pix suggests effectiveness in data-augmentation applications.
人工智能（AI）在医学影像中的应用仍然面临着难以收集和使用大型数据集的问题。一种解决这个问题的方法是使用生成对抗网络（GAN）生成的虚构图像进行数据扩充。然而，由于生成图像的质量和多样性，GAN 作为数据扩充技术尚未得到应用。为了通过生成多样化的图像来促进此类应用，本研究旨在使用基于 pix2pix 的模型从肿瘤草图生成自由形式的病变图像，该模型是一种源自 GAN 的图像到图像的翻译模型。由于 pix2pix 假设一对一的图像生成，因此不适合数据扩充，因此我们提出了 StylePix2pix，它是独立改进的，可以实现一对多的图像生成。所提出的模型引入了来自 StyleGAN 的映射网络和样式块。基于医生创建的 20 个肿瘤草图的图像生成结果表明，该方法可以再现具有复杂形状的肿瘤。此外，StylePix2pix 的一对多图像生成表明它在数据扩充应用中具有有效性。
