Department of Psychiatry, Gil Medical Center, Gachon University College of Medicine, Incheon, 21565, Republic of Korea.
Department of Psychiatry, Seoul National University College of Medicine, Seoul, 03080, Republic of Korea.
Mol Diagn Ther. 2018 Aug;22(4):475-483. doi: 10.1007/s40291-018-0334-z.
Fluorodeoxyglucose (FDG) positron emission tomography (PET) is useful to predict Alzheimer's disease (AD) conversion in patients with mild cognitive impairment (MCI). However, few studies have examined the extent to which FDG PET alone can predict AD conversion and compared the efficacy between visual and computer-assisted analysis directly.
The current study aimed to evaluate the value of FDG PET in predicting the conversion to AD in patients with MCI and to compare the predictive values of visual reading and computer-assisted analysis.
A total of 54 patients with MCI were evaluated with FDG PET and followed-up for 2 years with final diagnostic evaluation. FDG PET images were evaluated by (1) traditional visual rating, (2) composite score of visual rating of the brain cortices, and (3) composite score of computer-assisted analysis. Receiver operating characteristics (ROC) curves were compared to analyze predictive values.
Nineteen patients (35.2%) converted to AD from MCI. The area under the curve (AUC) of the ROC curve of the traditional visual rating, composite score of visual rating, and computer-assisted analysis were 0.67, 0.76, and 0.79, respectively. ROC curves of the composite scores of the visual rating and computer-assisted analysis were comparable (Z = 0.463, p = 0.643).
Visual rating and computer-assisted analysis of FDG PET scans were analogously accurate in predicting AD conversion in patients with MCI. Therefore, FDG PET may be a useful tool for screening AD conversion in patients with MCI, when using composite score, regardless of the method of interpretation.
氟代脱氧葡萄糖（FDG）正电子发射断层扫描（PET）可用于预测轻度认知障碍（MCI）患者的阿尔茨海默病（AD）转化。然而，很少有研究检查 FDG PET 单独预测 AD 转化的程度，并直接比较视觉和计算机辅助分析的效果。
本研究旨在评估 FDG PET 在预测 MCI 患者向 AD 转化中的价值，并比较视觉阅读和计算机辅助分析的预测值。
对 54 例 MCI 患者进行 FDG PET 评估，并进行 2 年的随访，最终进行诊断评估。FDG PET 图像通过（1）传统的视觉评分，（2）大脑皮质的视觉评分综合评分，和（3）计算机辅助分析的综合评分进行评估。比较受试者工作特征（ROC）曲线以分析预测值。
19 例患者（35.2%）从 MCI 转化为 AD。传统视觉评分、视觉评分综合评分和计算机辅助分析的 ROC 曲线下面积（AUC）分别为 0.67、0.76 和 0.79。视觉评分综合评分和计算机辅助分析的 ROC 曲线相当（Z = 0.463，p = 0.643）。
在预测 MCI 患者 AD 转化方面，FDG PET 的视觉评分和计算机辅助分析具有相似的准确性。因此，当使用综合评分时，FDG PET 可能是筛选 MCI 患者 AD 转化的有用工具，而与解释方法无关。
