利用深度学习解读多巴胺转运体显像用于帕金森综合征的鉴别诊断
本研究纳入1017例接受[11C]CFT多巴胺转运体PET显像的受试者(43例健康人、974例帕金森病患者,含特发性帕金森病、多系统萎缩或进行性核上性麻痹),构建3D深度卷积神经网络学习可区分特征以鉴别帕金森综合征,并采用全梯度显著图及深度学习引导的影像组学与定量分析进行解读。结果显示该网络在交叉验证和盲测中对特发性帕金森病、多系统萎缩和进行性核上性麻痹的鉴别具有较高的曲线下面积和敏感度、特异度,深度学习引导的结合比值在三组间差异有统计学意义,而传统壳核与尾状核结合比值在特发性帕金森病与多系统萎缩间无显著差异。作者认为该网络可解读多巴胺转运体显像中的深层信息,并显示出辅助帕金森综合征鉴别诊断的潜力。
要点
- 研究纳入1017例受试者,含43例健康人和974例帕金森病患者(特发性帕金森病、多系统萎缩、进行性核上性麻痹)。
- 3D深度卷积神经网络在交叉验证中鉴别特发性帕金森病、多系统萎缩、进行性核上性麻痹的曲线下面积分别为0.953、0.948、0.900。
- 盲测中三者敏感度分别为90.7%、84.1%、78.6%,特异度分别为88.4%、97.5%、93.3%。
核心结论
- 作者认为所开发的深度神经网络能够解读多巴胺转运体显像中的深层信息,并显示出辅助帕金森综合征鉴别诊断的潜力。
- 支持诊断决策的功能区域与已知帕金森综合征病理大体一致,并为特征选择和定量分析提供了更特异的指导。
研究设计与样本
- 本研究纳入1017例接受[11C]CFT多巴胺转运体PET显像的受试者,包括43例健康受试者和974例帕金森病患者(特发性帕金森病、多系统萎缩或进行性核上性麻痹)。
- 研究开发了3D深度卷积神经网络以学习可区分的多巴胺转运体特征,用于帕金森综合征的鉴别诊断。
- 采用全梯度显著图方法探究与网络决策机制相关的功能基础,并将深度学习引导的影像组学特征和定量分析与传统对应方法进行比较。
主要发现
- 所提出的网络在交叉验证中鉴别特发性帕金森病、多系统萎缩和进行性核上性麻痹的曲线下面积分别为0.953(敏感度87.7%,特异度93.2%)、0.948(敏感度93.7%,特异度97.5%)和0.900(敏感度81.5%,特异度93.7%)。
- 盲测中鉴别特发性帕金森病、多系统萎缩和进行性核上性麻痹的敏感度分别为90.7%、84.1%、78.6%,特异度分别为88.4%、97.5%、93.3%。
- 显著图显示决定诊断的最主要贡献区域位于帕金森综合征相关脑区,如壳核、尾状核和中脑。
- 深度学习引导的结合比值在特发性帕金森病、多系统萎缩和进行性核上性麻痹组间差异有统计学意义(P < 0.001),而传统壳核和尾状核结合比值在特发性帕金森病与多系统萎缩组间无显著差异(分别为P = 0.24和P = 0.30)。
- 与传统影像组学特征相比,深度学习引导的影像组学特征中在特发性帕金森病、多系统萎缩和进行性核上性麻痹组间差异有统计学意义的特征平均多出78.1%以上。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
PURPOSE: This work attempts to decode the discriminative information in dopamine transporter (DAT) imaging using deep learning for the differential diagnosis of parkinsonism. METHODS: This study involved 1017 subjects who underwent DAT PET imaging ([11C]CFT) including 43 healthy subjects and 974 parkinsonian patients with idiopathic Parkinson's disease (IPD), multiple system atrophy (MSA) or progressive supranuclear palsy (PSP). We developed a 3D deep convolutional neural network to learn distinguishable DAT features for the differential diagnosis of parkinsonism. A full-gradient saliency map approach was employed to investigate the functional basis related to the decision mechanism of the network. Furthermore, deep-learning-guided radiomics features and quantitative analysis were compared with their conventional counterparts to further interpret the performance of deep learning. RESULTS: The proposed network achieved area under the curve of 0.953 (sensitivity 87.7%, specificity 93.2%), 0.948 (sensitivity 93.7%, specificity 97.5%), and 0.900 (sensitivity 81.5%, specificity 93.7%) in the cross-validation, together with sensitivity of 90.7%, 84.1%, 78.6% and specificity of 88.4%, 97.5% 93.3% in the blind test for the differential diagnosis of IPD, MSA and PSP, respectively. The saliency map demonstrated the most contributed areas determining the diagnosis located at parkinsonism-related regions, e.g., putamen, caudate and midbrain. The deep-learning-guided binding ratios showed significant differences among IPD, MSA and PSP groups (P < 0.001), while the conventional putamen and caudate binding ratios had no significant difference between IPD and MSA (P = 0.24 and P = 0.30). Furthermore, compared to conventional radiomics features, there existed average above 78.1% more deep-learning-guided radiomics features that had significant differences among IPD, MSA and PSP. CONCLUSION: This study suggested the developed deep neural network can decode in-depth information from DAT and showed potential to assist the differential diagnosis of parkinsonism. The functional regions supporting the diagnosis decision were generally consistent with known parkinsonian pathology but provided more specific guidance for feature selection and quantitative analysis.
引用
- Zhao Y, Wu P, Wu J, et al. Decoding the dopamine transporter imaging for the differential diagnosis of parkinsonism using deep learning. Eur J Nucl Med Mol Imaging. 2022. PMID: 35588012. doi:10.1007/s00259-022-05804-x
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。