Department of Radiation Oncology (MAASTRO), GROW School for Oncology and Developmental Biology, Maastricht University Medical Centre+, Maastricht, 6229 ET, The Netherlands.
Department of Radiation Oncology, Radboud Institute for Health Sciences, Radboud University Medical Center, Nijmegen, 6525 GA, The Netherlands.
Phys Med Biol. 2021 Aug 3;66(16). doi: 10.1088/1361-6560/ac16c0.
Radiomics is an active area of research in medical image analysis, however poor reproducibility of radiomics has hampered its application in clinical practice. This issue is especially prominent when radiomic features are calculated from noisy images, such as low dose computed tomography (CT) scans. In this article, we investigate the possibility of improving the reproducibility of radiomic features calculated on noisy CTs by using generative models for denoising. Our work concerns two types of generative models-encoder-decoder network (EDN) and conditional generative adversarial network (CGAN). We then compared their performance against a more traditional 'non-local means' denoising algorithm. We added noise to sinograms of full dose CTs to mimic low dose CTs with two levels of noise: low-noise CT and high-noise CT. Models were trained on high-noise CTs and used to denoise low-noise CTs without re-training. We tested the performance of our model in real data, using a dataset of same-day repeated low dose CTs in order to assess the reproducibility of radiomic features in denoised images. EDN and the CGAN achieved similar improvements on the concordance correlation coefficients (CCC) of radiomic features for low-noise images from 0.87 [95%CI, (0.833, 0.901)] to 0.92 [95%CI, (0.909, 0.935)] and for high-noise images from 0.68 [95%CI, (0.617, 0.745)] to 0.92 [95%CI, (0.909, 0.936)], respectively. The EDN and the CGAN improved the test-retest reliability of radiomic features (mean CCC increased from 0.89 [95%CI, (0.881, 0.914)] to 0.94 [95%CI, (0.927, 0.951)]) based on real low dose CTs. These results show that denoising using EDN and CGANs could be used to improve the reproducibility of radiomic features calculated from noisy CTs. Moreover, images at different noise levels can be denoised to improve the reproducibility using the above models without need for re-training, provided the noise intensity is not excessively greater that of the high-noise CTs. To the authors' knowledge, this is the first effort to improve the reproducibility of radiomic features calculated on low dose CT scans by applying generative models.
放射组学是医学图像分析中的一个活跃研究领域，然而放射组学的可重复性差阻碍了其在临床实践中的应用。当从噪声图像（例如低剂量计算机断层扫描（CT）扫描）计算放射组学特征时，这个问题尤其突出。在本文中，我们研究了使用用于去噪的生成模型来提高从噪声 CT 计算出的放射组学特征的可重复性的可能性。我们的工作涉及两种类型的生成模型-编码器-解码器网络（EDN）和条件生成对抗网络（CGAN）。然后，我们将它们的性能与更传统的“非局部均值”去噪算法进行了比较。我们向全剂量 CT 的正弦图中添加噪声，以模拟两种噪声水平的低剂量 CT：低噪声 CT 和高噪声 CT。在高噪声 CT 上训练模型，并在不重新训练的情况下用于去噪低噪声 CT。我们在真实数据中测试了我们的模型的性能，使用同一天重复的低剂量 CT 的数据集来评估去噪图像中放射组学特征的可重复性。EDN 和 CGAN 分别将低噪声图像的放射组学特征的一致性相关系数（CCC）从 0.87 [95％CI，（0.833，0.901）]提高到 0.92 [95％CI，（0.909，0.935）]和高噪声图像的 CCC 从 0.68 [95％CI，（0.617，0.745）]提高到 0.92 [95％CI，（0.909，0.936）]。EDN 和 CGAN 提高了放射组学特征的测试-再测试可靠性（平均 CCC 从 0.89 [95％CI，（0.881，0.914）]提高到 0.94 [95％CI，（0.927，0.951）]）基于真实的低剂量 CT。这些结果表明，使用 EDN 和 CGAN 进行去噪可以用于提高从噪声 CT 计算出的放射组学特征的可重复性。此外，可以在不同噪声水平下对图像进行去噪，以使用上述模型提高可重复性，而无需重新训练，只要噪声强度不超过高噪声 CT 的噪声强度即可。据作者所知，这是首次通过应用生成模型来提高低剂量 CT 扫描上计算出的放射组学特征的可重复性。
