Department of Medical Physics and Biomedical engineering, Tehran University of Medical Sciences, Tehran, Iran.
Research Center for Molecular and Celular Imaging, Tehran University of Medical Sciences, Tehran, Iran.
Med Phys. 2022 Jun;49(6):3783-3796. doi: 10.1002/mp.15615. Epub 2022 Apr 11.
This study is aimed at examining the synergistic impact of motion and acquisition/reconstruction parameters on F-FDG PET image radiomic features in non-small cell lung cancer (NSCLC) patients, and investigating the robustness of features performance in differentiating NSCLC histopathology subtypes.
An in-house developed thoracic phantom incorporating lesions with different sizes was used with different reconstruction settings, including various reconstruction algorithms, number of subsets and iterations, full-width at half-maximum of post-reconstruction smoothing filter and acquisition parameters, including injected activity and test-retest with and without motion simulation. To simulate motion, a special motor was manufactured to simulate respiratory motion based on a normal patient in two directions. The lesions were delineated semi-automatically to extract 174 radiomic features. All radiomic features were categorized according to the coefficient of variation (COV) to select robust features. A cohort consisting of 40 NSCLC patients with adenocarcinoma (n = 20) and squamous cell carcinoma (n = 20) was retrospectively analyzed. Statistical analysis was performed to discriminate robust features in differentiating histopathology subtypes of NSCLC lesions.
Overall, 29% of radiomic features showed a COV ≤5% against motion. Forty-five percent and 76% of the features showed a COV ≤ 5% against the test-retest with and without motion in large lesions, respectively. Thirty-three percent and 45% of the features showed a COV ≤ 5% against different reconstruction parameters with and without motion, respectively. For NSCLC histopathological subtype differentiation, statistical analysis showed that 31 features were significant (p-value < 0.05). Two out of the 31 significant features, namely, the joint entropy of GLCM (AUC = 0.71, COV = 0.019) and median absolute deviation of intensity histogram (AUC = 0.7, COV = 0.046), were robust against the motion (same reconstruction setting).
Motion, acquisition, and reconstruction parameters significantly impact radiomic features, just as their synergies. Radiomic features with high predictive performance (statistically significant) in differentiating histopathological subtype of NSCLC may be eliminated due to non-reproducibility.
本研究旨在探讨运动和采集/重建参数对非小细胞肺癌（NSCLC）患者 18F-FDG PET 图像纹理特征的协同影响，并研究特征在区分 NSCLC 组织病理学亚型方面的稳健性。
使用带有不同大小病变的内部开发的胸部体模，并采用不同的重建设置，包括各种重建算法、子集和迭代的数量、后重建平滑滤波器的半峰全宽和采集参数，包括注入的活性以及有和没有运动模拟的测试-再测试。为了模拟运动，制造了一种特殊的电机，根据一名正常患者在两个方向上的运动来模拟呼吸运动。半自动勾勒病变以提取 174 个纹理特征。根据变异系数（COV）对所有纹理特征进行分类，以选择稳健的特征。回顾性分析了由 40 名 NSCLC 患者组成的队列，其中腺癌（n = 20）和鳞状细胞癌（n = 20）。对区分 NSCLC 病变组织病理学亚型的稳健特征进行了统计分析。
总体而言，29%的纹理特征的 COV 对运动的≤5%。大病变中，分别有 45%和 76%的特征对有和没有运动的测试-再测试的 COV 为≤5%。分别有 33%和 45%的特征对有和没有运动的不同重建参数的 COV 为≤5%。对于 NSCLC 组织病理学亚型的区分，统计分析显示有 31 个特征具有统计学意义（p 值<0.05）。在这 31 个显著特征中，有两个特征即 GLCM 的联合熵（AUC = 0.71，COV = 0.019）和强度直方图的中值绝对偏差（AUC = 0.7，COV = 0.046）是稳健的，不受运动影响（采用相同的重建设置）。
运动、采集和重建参数对纹理特征有显著影响，就像它们的协同作用一样。在区分 NSCLC 组织病理学亚型方面具有高预测性能（统计学上显著）的纹理特征可能由于不可再现性而被消除。
