We aimed to investigate the effects of ¹⁸F-FDG PET voxel intensity normalization on radiomic features of oropharyngeal squamous cell carcinoma (OPSCC) and machine learning–generated radiomic biomarkers. Methods: We extracted 1,037 ¹⁸F-FDG PET radiomic features quantifying the shape, intensity, and texture of 430 OPSCC primary tumors. The reproducibility of individual features across 3 intensity-normalized images (body-weight SUV, reference tissue activity ratio to lentiform nucleus of brain and cerebellum) and the raw PET data was assessed using an intraclass correlation coefficient (ICC). We investigated the effects of intensity normalization on the features’ utility in predicting the human papillomavirus (HPV) status of OPSCCs in univariate logistic regression, receiver-operating-characteristic analysis, and extreme-gradient-boosting (XGBoost) machine-learning classifiers. Results: Of 1,037 features, a high (ICC ≥ 0.90), medium (0.90 > ICC ≥ 0.75), and low (ICC < 0.75) degree of reproducibility across normalization methods was attained in 356 (34.3%), 608 (58.6%), and 73 (7%) features, respectively. In univariate analysis, features from the PET normalized to the lentiform nucleus had the strongest association with HPV status, with 865 of 1,037 (83.4%) significant features after multiple testing corrections and a median area under the receiver-operating-characteristic curve (AUC) of 0.65 (interquartile range, 0.62–0.68). Similar tendencies were observed in XGBoost models, with the lentiform nucleus–normalized model achieving the numerically highest average AUC of 0.72 (SD, 0.07) in the cross validation within the training cohort. The model generalized well to the validation cohorts, attaining an AUC of 0.73 (95% CI, 0.60–0.85) in independent validation and 0.76 (95% CI, 0.58–0.95) in external validation. The AUCs of the XGBoost models were not significantly different. Conclusion: Only one third of the features demonstrated a high degree of reproducibility across intensity-normalization techniques, making uniform normalization a prerequisite for interindividual comparability of radiomic markers. The choice of normalization technique may affect the radiomic features’ predictive value with respect to HPV. Our results show trends that normalization to the lentiform nucleus may improve model performance, although more evidence is needed to draw a firm conclusion.

^{我们的目的是研究18} F-FDG PET 体素强度标准化对口咽鳞状细胞癌 (OPSCC) 放射组学特征和机器学习生成的放射组学生物标志物的影响。方法：我们提取了 1,037 个¹⁸ F-FDG PET 放射组学特征，量化了 430 个 OPSCC 原发肿瘤的形状、强度和纹理。使用组内相关系数 (ICC) 评估 3 个强度归一化图像（体重 SUV、参考组织与大脑和小脑豆状核的活动比）和原始 PET 数据中个体特征的再现性。我们研究了强度归一化对单变量逻辑回归、接受者操作特征分析和极限梯度提升 (XGBoost) 机器学习分类器中预测 OPSCC 人乳头瘤病毒 (HPV) 状态的特征效用的影响。结果：在 1,037 个特征中，356 个（34.3%）、608 个（58.6%）在标准化方法中获得了高（ICC ≥ 0.90）、中（0.90 > ICC ≥ 0.75）和低（ICC < 0.75）程度的可重复性、 和 73 (7%) 个特征。在单变量分析中，归一化为豆状核的 PET 特征与 HPV 状态的关联性最强，经过多次测试校正后，1,037 个特征中有 865 个 (83.4%) 具有显着特征，并且受试者工作特征曲线下面积 (AUC) 的中位面积0.65（四分位数范围，0.62–0.68）。在 XGBoost 模型中也观察到了类似的趋势，在训练队列内的交叉验证中，豆状核归一化模型实现了数值最高的平均 AUC 0.72（SD，0.07）。该模型很好地推广到了验证队列，在独立验证中获得了 0.73（95% CI，0.60-0.85）的 AUC，在外部验证中获得了 0.76（95% CI，0.58-0.95）。 XGBoost 模型的 AUC 没有显着差异。结论：只有三分之一的特征表现出强度归一化技术的高度可重复性，使得统一归一化成为放射组学标记个体间可比性的先决条件。标准化技术的选择可能会影响放射组学特征对 HPV 的预测价值。我们的结果表明，豆状核标准化可能会提高模型性能，但还需要更多证据来得出明确的结论。