论文标题
使用生成对抗网络进行CT标准化及其对放射线特征的影响
Using a Generative Adversarial Network for CT Normalization and its Impact on Radiomic Features
论文作者
论文摘要
计算机辅助诊断（CADX）系统可以使用形态和基于纹理（放射线）特征在胸部CT扫描上识别和分类胸部CT扫描的潜在恶性肺结核。但是，由于剂量水平和切片厚度的变化，放射素特征对采集的差异敏感。这项研究研究了从异质CT扫描作为输入中产生归一化扫描的可行性。我们从40次低剂量胸部CT扫描中获得了投影数据，以10％，25％和50％的剂量模拟采集，并以1.0mm和2.0mm的切片厚度重建扫描。使用3D生成对抗网络（GAN）同时将减少剂量，厚切片（2.0mm）图像归一化为正常剂量（100％），较薄的切片（1.0mm）图像。我们使用峰值信号噪声比（PSNR），结构相似性指数（SSIM）和学习的知觉图像贴片相似性（LPIPS）评估了归一化图像质量。与基线CNN方法相比，我们的GAN将感知相似性提高了35％。我们的分析还表明，基于GAN的方法导致九个研究的放射线特征的误差明显较小（P值<0.05）。这些结果表明，GAN可用于使异质CT图像归一化并降低放射线特征值的变异性。
Computer-Aided-Diagnosis (CADx) systems assist radiologists with identifying and classifying potentially malignant pulmonary nodules on chest CT scans using morphology and texture-based (radiomic) features. However, radiomic features are sensitive to differences in acquisitions due to variations in dose levels and slice thickness. This study investigates the feasibility of generating a normalized scan from heterogeneous CT scans as input. We obtained projection data from 40 low-dose chest CT scans, simulating acquisitions at 10%, 25% and 50% dose and reconstructing the scans at 1.0mm and 2.0mm slice thickness. A 3D generative adversarial network (GAN) was used to simultaneously normalize reduced dose, thick slice (2.0mm) images to normal dose (100%), thinner slice (1.0mm) images. We evaluated the normalized image quality using peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) and Learned Perceptual Image Patch Similarity (LPIPS). Our GAN improved perceptual similarity by 35%, compared to a baseline CNN method. Our analysis also shows that the GAN-based approach led to a significantly smaller error (p-value < 0.05) in nine studied radiomic features. These results indicated that GANs could be used to normalize heterogeneous CT images and reduce the variability in radiomic feature values.
