Department of Nuclear Medicine, Centre Antoine Lacassagne, Université Côte d'Azur, Nice, France.
Laboratoire TIRO UMR E4320, Université Côte d'Azur, Nice, France.
Eur J Nucl Med Mol Imaging. 2022 Sep;49(11):3787-3796. doi: 10.1007/s00259-022-05816-7. Epub 2022 May 14.
FDOPA PET shows good performance for the diagnosis of striatal dopaminergic denervation, making it a valuable tool for the differential diagnosis of Parkinsonism. Textural features are image biomarkers that could potentially improve the early diagnosis and monitoring of neurodegenerative parkinsonian syndromes. We explored the performances of textural features for binary classification of FDOPA scans.
We used two FDOPA PET datasets: 443 scans for feature selection, and 100 scans from a different PET/CT system for model testing. Scans were labelled according to expert interpretation (dopaminergic denervation versus no dopaminergic denervation). We built LASSO logistic regression models using 43 biomarkers including 32 textural features. Clinical data were also collected using a shortened UPDRS scale.
The model built from the clinical data alone had a mean area under the receiver operating characteristics (AUROC) of 63.91. Conventional imaging features reached a maximum score of 93.47 but the addition of textural features significantly improved the AUROC to 95.73 (p < 0.001), and 96.10 (p < 0.001) when limiting the model to the top three features: GLCM_Correlation, Skewness and Compacity. Testing the model on the external dataset yielded an AUROC of 96.00, with 95% sensitivity and 97% specificity. GLCM_Correlation was one of the most independent features on correlation analysis, and systematically had the heaviest weight in the classification model.
A simple model with three radiomic features can identify pathologic FDOPA PET scans with excellent sensitivity and specificity. Textural features show promise for the diagnosis of parkinsonian syndromes.
FDOPA PET 对纹状体多巴胺能神经变性的诊断具有良好的性能，使其成为帕金森病鉴别诊断的有价值的工具。纹理特征是图像生物标志物，可能有助于早期诊断和监测神经退行性帕金森综合征。我们探讨了纹理特征在 FDOPA 扫描的二进制分类中的性能。
我们使用了两个 FDOPA PET 数据集：443 个扫描用于特征选择，以及来自不同 PET/CT 系统的 100 个扫描用于模型测试。扫描根据专家解释（多巴胺能神经变性与无多巴胺能神经变性）进行标记。我们使用包括 32 个纹理特征的 43 个生物标志物构建了 LASSO 逻辑回归模型。还使用缩短的 UPDRS 量表收集了临床数据。
仅基于临床数据构建的模型的接收器操作特性（AUROC）的平均值为 63.91。常规成像特征的最高得分为 93.47，但纹理特征的加入显著提高了 AUROC 至 95.73（p < 0.001），当将模型限制为前三个特征时，AUROC 为 96.10（p < 0.001）：GLCM_Correlation、偏度和紧凑度。在外部数据集上测试模型时，AUROC 为 96.00，灵敏度为 95%，特异性为 97%。在相关分析中，GLCM_Correlation 是最独立的特征之一，并且在分类模型中系统地具有最重的权重。
具有三个放射组学特征的简单模型可以识别具有出色敏感性和特异性的病理性 FDOPA PET 扫描。纹理特征有望用于帕金森综合征的诊断。
