Image synthesis with deep convolutional generative adversarial networks for material decomposition in dual-energy CT from a kilovoltage CT
Image synthesis with deep convolutional generative adversarial networks for material decomposition in dual-energy CT from a kilovoltage CT
复制标题
DOI：
10.1016/j.compbiomed.2020.104111
复制
发表时间：
2020-11
影响因子：
7.7
通讯作者：
Daisuke Kawahara;A. Saito;S. Ozawa;Y. Nagata
中科院分区：
文献类型：
--
作者：
Daisuke Kawahara;A. Saito;S. Ozawa;Y. Nagata
生成对抗网络（Generative Adversarial Networks，GANs）已经得到了广泛的应用，并有望应用于临床检查和成像领域。本研究的目的是合成从双能量计算机断层扫描（DECT）重建的骨-水（骨（水））和脂肪-水（脂肪（水））的材料分解图像，使用等效千伏-CT（kV-CT）图像和深度条件GAN。有效原子序数图像用DECT重建。我们使用了28名患者的18，084张图像，分为两个数据集：模型的训练数据包括16，146张图像（20名患者），用于评估的测试数据包括1938张图像（8名患者）。建立了120 kVp下等效单能CT图像与有效原子序数图像的图像预测框架。图像合成框架基于具有生成器和卷积神经网络的CNN。评价了平均绝对误差（MAE）、相对均方误差（MSE）、相对均方根误差（RMSE）、峰值信噪比（PSNR）、结构相似性指数（SSIM）和互信息（MI）。骨（水）和脂肪（水）的合成和参考材料分解图像之间的Hounsfield单位（HU）差异分别在5.3 HU和20.3 HU范围内。骨（水）图像的合成和参考材料分解的平均MAE、MSE、RMSE、SSIM和MI分别为0.8、1.3、0.9、0.9、55.3和0.8。脂肪（水）图像的合成和参考材料分解的平均MAE、MSE、RMSE、SSIM和MI分别为0.0、0.0、0.1、0.9、72.1和1.4。所提出的模型可以作为一个合适的替代现有的方法重建的材料分解图像的骨（水）和脂肪（水）重建通过DECT从千伏CT。
Generative Adversarial Networks (GANs) have been widely used and it is expected to use for the clinical examination and image. The objective of the current study was to synthesize material decomposition images of bone-water (bone(water)) and fat-water (fat(water)) reconstructed from dual-energy computed tomography (DECT) using an equivalent kilovoltage-CT (kV-CT) image and a deep conditional GAN. The effective atomic number images were reconstructed using DECT. We used 18,084 images of 28 patients divided into two datasets: the training data for the model included 16,146 images (20 patients) and the test data for evaluation included 1938 images (8 patients). Image prediction frameworks of the equivalent single energy CT images at 120 kVp to the effective atomic number images were created. The image-synthesis framework was based on a CNN with a generator and discriminator. The mean absolute error (MAE), relative mean square error (MSE), relative root mean square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and mutual information (MI) were evaluated. The Hounsfield unit (HU) difference between the synthesized and reference material decomposition images of bone(water) and fat(water) were within 5.3 HU and 20.3 HU, respectively. The average MAE, MSE, RMSE, SSIM, and MI of the synthesized and reference material decomposition of the bone(water) images were 0.8, 1.3, 0.9, 0.9, 55.3, and 0.8, respectively. The average MAE, MSE, RMSE, SSIM, and MI of the synthesized and reference material decomposition of the fat(water) images were 0.0, 0.0, 0.1, 0.9, 72.1, and 1.4, respectively. The proposed model can act as a suitable alternative to the existing methods for the reconstruction of material decomposition images of bone(water) and fat(water) reconstructed via DECT from kV-CT.