⁶⁸ Ga-DOTATATE PET-CT is routinely used for imaging neuroendocrine tumor (NET) somatostatin receptor subtype 2 (SSTR2) density in patients, and is complementary to FDG PET-CT for improving the accuracy of NET detection, characterization, grading, staging, and predicting/monitoring NET responses to treatment. Performing sequential ¹⁸ F-FDG and ⁶⁸ Ga-DOTATATE PET scans would require 2 or more days and can delay patient care. To align temporal and spatial measurements of ¹⁸ F-FDG and ⁶⁸ Ga-DOTATATE PET, and to reduce scan time and CT radiation exposure to patients, we propose a single-imaging session dual-tracer dynamic PET acquisition protocol in the study. A recurrent extreme gradient boosting (rXGBoost) machine learning algorithm was proposed to separate the mixed ¹⁸ F-FDG and ⁶⁸ Ga-DOTATATE time activity curves (TACs) for the region of interest (ROI) based quantification with tracer kinetic modeling. A conventional parallel multi-tracer compartment modeling method was also implemented for reference. Single-scan dual-tracer dynamic PET was simulated from 12 NET patient studies with ¹⁸ F-FDG and ⁶⁸ Ga-DOTATATE 45-min dynamic PET scans separately obtained within 2 days. Our experimental results suggested an ¹⁸ F-FDG injection first followed by ⁶⁸ Ga-DOTATATE with a minimum 5 min delayed injection protocol for the separation of mixed ¹⁸ F-FDG and ⁶⁸ Ga-DOTATATE TACs using rXGBoost algorithm followed by tracer kinetic modeling is highly feasible.

中文翻译：

---

基于机器学习的单成像会话双示踪剂 18F-FDG 和 68Ga-DOTATATE 动态 PET-CT 在肿瘤学中的无创定量


⁶⁸ Ga-DOTATATE PET-CT 常规用于患者神经内分泌肿瘤 (NET) 生长抑素受体亚型 2 (SSTR2) 密度成像，与 FDG PET-CT 互补，可提高 NET 检测、表征、分级、分期的准确性。预测/监测 NET 对治疗的反应。连续执行¹⁸ F-FDG 和⁶⁸ Ga-DOTATATE PET 扫描需要 2 天或更长时间，并且可能会延误患者护理。为了协调¹⁸ F-FDG 和⁶⁸ Ga-DOTATATE PET 的时间和空间测量，并减少患者的扫描时间和 CT 辐射暴露，我们在研究中提出了单成像会话双示踪剂动态 PET 采集协议。提出了一种循环极端梯度增强 (rXGBoost) 机器学习算法，用于分离混合的¹⁸ F-FDG 和⁶⁸ Ga-DOTATATE 时间活度曲线 (TAC)，用于基于感兴趣区域 (ROI) 的量化和示踪动力学建模。还实现了传统的并行多示踪剂室建模方法以供参考。单扫描双示踪剂动态 PET 是根据 12 名 NET 患者研究进行模拟的，其中使用 2 天内分别获得的¹⁸ F-FDG 和⁶⁸ Ga-DOTATATE 45 分钟动态 PET 扫描。我们的实验结果表明，首先注射¹⁸ F-FDG，然后注射⁶⁸ Ga-DOTATATE，延迟注射至少 5 分钟，使用 rXGBoost 算法分离混合¹⁸ F-FDG 和⁶⁸ Ga-DOTATATE TAC，然后进行示踪动力学建模，效果非常好。可行的。