基于影像组学的深度神经网络提升可解释性:18F-FDG PET 影像分类帕金森综合征的概念验证
本双中心研究分析华山医院 1275 例特发性帕金森病(IPD)、多系统萎缩(MSA)、进行性核上性麻痹(PSP)患者及 223 例健康对照,以及慕尼黑大学医院 90 例 IPD、MSA、PSP 患者(34 例 IPD、17 例 MSA、39 例 PSP)的 18F-FDG PET 影像,提出影像组学引导的双通道深度神经网络(RDDNN)。模型在内部盲测队列 AUC=0.99、准确率=0.98,外部队列 AUC=0.94、准确率=0.81,评估时间较核医学专家阅片显著缩短(p < 0.001)。
要点
- RDDNN 在内部盲测队列达 AUC=0.99、准确率=0.98,外部队列 AUC=0.94、准确率=0.81。
- 模型评估时间较核医学专家阅片显著缩短(p < 0.001)。
- 热图显示 IPD、MSA、PSP 在解剖相关区域呈疾病特异性激活。
核心结论
- RDDNN 框架为帕金森综合征分类提供了临床可解释且可重复的深度神经网络方案。
- 该框架整合影像组学与注意力建模,增强病灶定位、支持临床决策,性能与人类专家相当,同时大幅提升诊断效率。
研究设计与样本
- 双中心研究,再分析中国与德国已充分研究的帕金森病患者 18F-FDG PET 影像队列。
- 华山医院队列含 1275 例 IPD、MSA、PSP 患者及 223 例健康对照;慕尼黑大学医院队列含 90 例 IPD、MSA、PSP 患者(34 例 IPD、17 例 MSA、39 例 PSP)。
- FDG 扫描为注射后 60 分钟 10 分钟静态采集,并以全脑活性归一化。
- RDDNN 结合膨胀卷积网络提取的局部特征与基于 Transformer 自注意力网络的全局特征。
- 模型性能以分类指标评估,并与影像组学及深度神经网络方法比较;输出还与核医学专家视觉评估比较以评估可解释性与时间效率。
主要发现
- 内部盲测队列中 RDDNN 达 AUC=0.99、准确率=0.98。
- SHAP 与相关性分析共同提示通道间存在互补信息,部分具有临床可解释性。
- 外部队列中模型保持稳健性能(AUC=0.94、准确率=0.81),特征模式跨人群一致。
- 模型评估时间较核医学专家阅片显著缩短(p < 0.001)。
- 热图显示 IPD、MSA、PSP 在解剖相关区域呈疾病特异性激活。
关键数值与推荐
- 内部盲测队列:AUC=0.99、准确率=0.98。
- 外部队列:AUC=0.94、准确率=0.81。
- 评估时间较核医学专家阅片显著缩短,p < 0.001。
临床怎么用
- 原文结论认为 RDDNN 框架为帕金森综合征分类提供临床可解释且可重复的深度神经网络方案。
- 原文结论认为该框架增强病灶定位、支持临床决策,性能与人类专家相当,并大幅提升诊断效率。
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
OBJECTIVE: Interpretability and reproducibility remain major challenges in applying deep neural network (DNN) to neuroimaging-based diagnosis. This study proposes a radiomics-guided dual-channel deep neural network (RDDNN) to improve feature transparency and enhance clinical understanding in the classification of Parkinsonian syndromes. METHODS: In this bi-centric study, we analysed two independent cohorts comprising 1,275 patients with idiopathic Parkinson's disease (IPD), multiple system atrophy (MSA), and progressive supranuclear palsy (PSP), alongside 223 healthy controls from Huashan Hospital and 90 patients with IPD, MSA, and PSP (34IPD, 17MSA, 39PSP) from the University Hospital Munich. It is a re-analysis of well-studied Chinese and German cohorts of 18F-fluorodeoxyglucose Positron emission tomography (FDG-PET) imaging of parkinsonian patients and the FDG scans were of 10-min static acquisition at 60 min post FDG injection and normalized against whole brain activity. The RDDNN model combines local features extracted via dilated convolutional networks and global features derived from Transformer-based self-attention networks. Model performance was evaluated using classification metrics and compared to radiomics and DNN approaches. The model's outputs were also compared with nuclear medicine specialists' visual assessments to assess interpretability and time efficiency. Furthermore, SHapley Additive Explanations (SHAP), Layer-wise Class Activation Mapping (Layer-CAM), and Rollout Attention Map (RAM) were employed to evaluate which features played the most critical roles in the model's final classification decisions after supervised training, and to examine how both networks spatially corresponded to known brain connectivity regions. RESULTS: In the internal blind-test cohort, the RDDNN achieved high accuracy (AUC = 0.99, accuracy = 0.98). SHAP and correlation analyses jointly indicated complementary information across channels, some of which were clinically interpretable. In the external cohort, the model maintained robust performance (AUC = 0.94, accuracy = 0.81), with consistent feature patterns across populations. The model significantly reduced evaluation time compared to nuclear medicine specialists' readings (p < 0.001), and the heatmaps showed disease-specific activation in anatomically relevant regions for IPD, MSA, and PSP. CONCLUSION: The RDDNN framework provides a clinically interpretable and reproducible DNN solution for classifying Parkinsonian disorders. By integrating radiomics and attention-based modeling, it enhances lesion localization, supports clinical decision-making, and offers performance comparable to human specialists-while substantially improving diagnostic efficiency.
引用
- Li C, Jiao F, Wu S, et al. Enhancing interpretability of AI with radiomics-based deep neural network: proof of concept in the classification of Parkinsonian syndromes with 18F-FDG PET imaging. Eur J Nucl Med Mol Imaging. 2025. PMID: 40932612. doi:10.1007/s00259-025-07478-7
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。