Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, GA 30322, United States of America.
Department of Biomedical Informatics, Emory University, Atlanta, GA 30322, United States of America.
Phys Med Biol. 2023 May 5;68(10):105004. doi: 10.1088/1361-6560/acca5c.
. Artificial intelligence (AI) methods have gained popularity in medical imaging research. The size and scope of the training image datasets needed for successful AI model deployment does not always have the desired scale. In this paper, we introduce a medical image synthesis framework aimed at addressing the challenge of limited training datasets for AI models.. The proposed 2D image synthesis framework is based on a diffusion model using a Swin-transformer-based network. This model consists of a forward Gaussian noise process and a reverse process using the transformer-based diffusion model for denoising. Training data includes four image datasets: chest x-rays, heart MRI, pelvic CT, and abdomen CT. We evaluated the authenticity, quality, and diversity of the synthetic images using visual Turing assessments conducted by three medical physicists, and four quantitative evaluations: the Inception score (IS), Fréchet Inception Distance score (FID), feature similarity and diversity score (DS, indicating diversity similarity) between the synthetic and true images. To leverage the framework value for training AI models, we conducted COVID-19 classification tasks using real images, synthetic images, and mixtures of both images.. Visual Turing assessments showed an average accuracy of 0.64 (accuracy converging to50%indicates a better realistic visual appearance of the synthetic images), sensitivity of 0.79, and specificity of 0.50. Average quantitative accuracy obtained from all datasets were IS = 2.28, FID = 37.27, FDS = 0.20, and DS = 0.86. For the COVID-19 classification task, the baseline network obtained an accuracy of 0.88 using a pure real dataset, 0.89 using a pure synthetic dataset, and 0.93 using a dataset mixed of real and synthetic data.. A image synthesis framework was demonstrated for medical image synthesis, which can generate high-quality medical images of different imaging modalities with the purpose of supplementing existing training sets for AI model deployment. This method has potential applications in many data-driven medical imaging research.
人工智能（AI）方法在医学影像研究中越来越受欢迎。成功部署 AI 模型所需的训练图像数据集的大小和范围并不总是具有理想的规模。在本文中，我们介绍了一种旨在解决 AI 模型训练数据集有限挑战的医学图像合成框架。
所提出的 2D 图像合成框架基于使用基于 Swin-Transformer 的网络的扩散模型。该模型由正向高斯噪声过程和使用基于 Transformer 的扩散模型进行去噪的反向过程组成。训练数据包括四个图像数据集：胸部 X 光片、心脏 MRI、盆腔 CT 和腹部 CT。我们使用三位医学物理学家进行的视觉图灵评估来评估合成图像的真实性、质量和多样性，并进行了四项定量评估：Inception 分数（IS）、Fréchet Inception 距离分数（FID）、特征相似性和多样性分数（DS，指示合成图像和真实图像之间的多样性相似性）。为了利用该框架为 AI 模型训练的价值，我们使用真实图像、合成图像和两者的混合图像进行了 COVID-19 分类任务。
视觉图灵评估显示平均准确率为 0.64（准确率收敛到 50%表示合成图像的真实视觉外观更好），敏感性为 0.79，特异性为 0.50。从所有数据集获得的平均定量准确率为 IS = 2.28、FID = 37.27、FDS = 0.20 和 DS = 0.86。对于 COVID-19 分类任务，基线网络使用纯真实数据集获得了 0.88 的准确率，使用纯合成数据集获得了 0.89 的准确率，使用真实和合成数据混合的数据集获得了 0.93 的准确率。
本文展示了一种用于医学图像合成的框架，该框架可以生成不同成像方式的高质量医学图像，旨在补充 AI 模型部署的现有训练集。该方法在许多数据驱动的医学影像研究中有潜在的应用。
