Translational Neuroimaging Laboratory, The McGill University Research Centre for Studies in Aging, Douglas Hospital, McGill University, Montréal, QC, Canada.
Department of Neurology and Neurosurgery, Faculty of Medicine, McGill University, Montréal, QC, Canada.
Ann Nucl Med. 2024 Jul;38(7):493-507. doi: 10.1007/s12149-024-01923-7. Epub 2024 Apr 4.
This study aimed to examine the robustness of positron emission tomography (PET) radiomic features extracted via different segmentation methods before and after ComBat harmonization in patients with non-small cell lung cancer (NSCLC).
We included 120 patients (positive recurrence = 46 and negative recurrence = 74) referred for PET scanning as a routine part of their care. All patients had a biopsy-proven NSCLC. Nine segmentation methods were applied to each image, including manual delineation, K-means (KM), watershed, fuzzy-C-mean, region-growing, local active contour (LAC), and iterative thresholding (IT) with 40, 45, and 50% thresholds. Diverse image discretizations, both without a filter and with different wavelet decompositions, were applied to PET images. Overall, 6741 radiomic features were extracted from each image (749 radiomic features from each segmented area). Non-parametric empirical Bayes (NPEB) ComBat harmonization was used to harmonize the features. Linear Support Vector Classifier (LinearSVC) with L1 regularization For feature selection and Support Vector Machine classifier (SVM) with fivefold nested cross-validation was performed using StratifiedKFold with 'n_splits' set to 5 to predict recurrence in NSCLC patients and assess the impact of ComBat harmonization on the outcome.
From 749 extracted radiomic features, 206 (27%) and 389 (51%) features showed excellent reliability (ICC ≥ 0.90) against segmentation method variation before and after NPEB ComBat harmonization, respectively. Among all, 39 features demonstrated poor reliability, which declined to 10 after ComBat harmonization. The 64 fixed bin widths (without any filter) and wavelets (LLL)-based radiomic features set achieved the best performance in terms of robustness against diverse segmentation techniques before and after ComBat harmonization. The first-order and GLRLM and also first-order and NGTDM feature families showed the largest number of robust features before and after ComBat harmonization, respectively. In terms of predicting recurrence in NSCLC, our findings indicate that using ComBat harmonization can significantly enhance machine learning outcomes, particularly improving the accuracy of watershed segmentation, which initially had fewer reliable features than manual contouring. Following the application of ComBat harmonization, the majority of cases saw substantial increase in sensitivity and specificity.
Radiomic features are vulnerable to different segmentation methods. ComBat harmonization might be considered a solution to overcome the poor reliability of radiomic features.
本研究旨在检查非小细胞肺癌（NSCLC）患者在进行 ComBat 均衡化前后，通过不同分割方法提取的正电子发射断层扫描（PET）放射组学特征的稳健性。
我们纳入了 120 名患者（阳性复发=46 例，阴性复发=74 例），他们接受 PET 扫描是其常规治疗的一部分。所有患者均经活检证实患有 NSCLC。对每个图像应用了 9 种分割方法，包括手动描绘、K-均值（KM）、分水岭、模糊 C-均值、区域生长、局部主动轮廓（LAC）和迭代阈值（IT），阈值分别为 40%、45%和 50%。对 PET 图像应用了不同的图像离散化方法，包括无滤波器和不同的小波分解。总体而言，从每个图像中提取了 6741 个放射组学特征（每个分割区域 749 个特征）。使用非参数经验贝叶斯（NPEB）ComBat 均衡化来均衡特征。使用带有 L1 正则化的线性支持向量分类器（LinearSVC）进行特征选择，并使用带有五重嵌套交叉验证的支持向量机（SVM）分类器，使用 StratifiedKFold 进行预测，将“n_splits”设置为 5，以预测 NSCLC 患者的复发情况，并评估 ComBat 均衡化对结果的影响。
在 749 个提取的放射组学特征中，有 206 个（27%）和 389 个（51%）特征在经过 NPEB ComBat 均衡化前后的分割方法变化方面表现出良好的可靠性（ICC≥0.90）。在所有特征中，有 39 个特征表现出较差的可靠性，经过 ComBat 均衡化后降至 10 个。在未经任何滤波器处理的 64 个固定 bin 宽度和基于小波（LLL）的放射组学特征集在经过 ComBat 均衡化前后的多种分割技术的稳健性方面表现出最佳性能。在经过 ComBat 均衡化前后，一阶和 GLRLM 以及一阶和 NGTDM 特征族分别表现出最大数量的稳健特征。在预测 NSCLC 复发方面，我们的研究结果表明，使用 ComBat 均衡化可以显著提高机器学习的结果，特别是可以提高分水岭分割的准确性，而分水岭分割最初的可靠特征比手动描绘的要少。在应用 ComBat 均衡化后，大多数情况下都显著提高了敏感性和特异性。
放射组学特征容易受到不同分割方法的影响。ComBat 均衡化可能是克服放射组学特征可靠性差的一种解决方案。
