Department of Functional Imaging and Artificial Intelligence, Kanazawa University Graduate School of Advanced Preventive Medical Sciences, 13-1 Takara-machi, Kanazawa, 920-8640, Japan.
Department of Nuclear Medicine, Kanazawa University, Kanazawa, Japan.
Ann Nucl Med. 2022 Aug;36(8):765-776. doi: 10.1007/s12149-022-01759-z. Epub 2022 Jul 7.
I-ioflupane has been clinically applied to dopamine transporter imaging and visual interpretation assisted by region-of-interest (ROI)-based parameters. We aimed to build a multivariable model incorporating machine learning (ML) that could accurately differentiate abnormal profiles on I-ioflupane images and diagnose Parkinson syndrome or disease and dementia with Lewy bodies (PS/PD/DLB).
We assessed I-ioflupane images from 239 patients with suspected neurodegenerative diseases or dementia and classified them as having PS/PD/DLB or non-PS/PD/DLB. The image features of high or low uptake (F1), symmetry or asymmetry (F2), and comma- or dot-like patterns of caudate and putamen uptake (F3) were analyzed on 137 images from one hospital for training. Direct judgement of normal or abnormal profiles (F4) was also examined. Machine learning methods included logistic regression (LR), k-nearest neighbors (kNNs), and gradient boosted trees (GBTs) that were assessed using fourfold cross-validation. We generated the following multivariable models for the test database (n = 102 from another hospital): Model 1, ROI-based measurements of specific binding ratios and asymmetry indices; Model 2, ML-based judgement of abnormalities (F4); and Model 3, features F1, F2 and F3, plus patient age. Diagnostic accuracy was compared using areas under receiver-operating characteristics curves (AUC).
The AUC was high with all ML methods (0.92-0.96) for high or low uptake. The AUC was the highest for symmetry or asymmetry with the kNN method (AUC 0.75) and the comma-dot feature with the GBT method (AUC 0.94). Based on the test data set, the diagnostic accuracy for a diagnosis of PS/PD/DLB was 0.86 ± 0.04 (SE), 0.87 ± 0.04, and 0.93 ± 0.02 for Models 1, 2 and 3, respectively. The AUC was optimal for Model 3, and significantly differed between Models 3 and 1 (p = 0.027), and 3 and 2 (p = 0.029).
Image features such as high or low uptake, symmetry or asymmetry, and comma- or dot-like profiles can be determined using ML. The diagnostic accuracy of differentiating PS/PD/DLB was the highest for the multivariate model with three features and age compared with the conventional ROI-based method.
碘[123I] 比曲替啶已被临床应用于多巴胺转运体成像，通过基于感兴趣区（ROI）的参数进行视觉解读。我们旨在建立一个包含机器学习（ML）的多变量模型，能够准确区分碘[123I]比曲替啶图像上的异常模式，并诊断帕金森综合征或疾病和路易体痴呆（PS/PD/DLB）。
我们评估了 239 名疑似神经退行性疾病或痴呆患者的碘[123I]比曲替啶图像，并将其分为 PS/PD/DLB 或非 PS/PD/DLB 组。我们对来自一家医院的 137 张图像进行了高或低摄取（F1）、对称或不对称（F2）、尾状核和壳核摄取的逗号或点状模式（F3）的图像特征分析。还检查了正常或异常轮廓的直接判断（F4）。机器学习方法包括逻辑回归（LR）、k-最近邻（kNNs）和梯度提升树（GBTs），它们使用四折交叉验证进行评估。我们为来自另一家医院的测试数据库（n=102）生成了以下多变量模型：模型 1，ROI 特定结合比和不对称指数的测量值；模型 2，异常的基于 ML 的判断（F4）；和模型 3，特征 F1、F2 和 F3，加患者年龄。使用接收器工作特征曲线下的面积（AUC）比较诊断准确性。
所有 ML 方法（0.92-0.96）对高或低摄取的 AUC 均较高。kNN 方法的 AUC 最高，用于对称或不对称（AUC 0.75），GBT 方法的 AUC 最高，用于逗号-点特征（AUC 0.94）。基于测试数据集，PS/PD/DLB 诊断的诊断准确性分别为模型 1、2 和 3 的 0.86±0.04（SE）、0.87±0.04 和 0.93±0.02。AUC 对于模型 3 最佳，且模型 3 与模型 1（p=0.027）和模型 3 与模型 2（p=0.029）之间存在显著差异。
可以使用 ML 确定高或低摄取、对称或不对称以及逗号或点状轮廓等图像特征。与传统的 ROI 基于方法相比，具有三个特征和年龄的多变量模型对区分 PS/PD/DLB 的诊断准确性最高。
