用脑科学先验知识提升脑电抑郁诊断准确率
ELPG-DTFS: Prior-Guided Adaptive Time-Frequency Graph Neural Network for EEG Depression Diagnosis
- 引入动态频域图结构和跨频带注意力机制
- 在128通道数据集上达97.63%准确率，超当前最佳
- 适合需要可解释性诊断模型的临床研究者
及时客观地筛查重度抑郁症（MDD）至关重要，但目前诊断仍依赖主观量表。脑电图（EEG）提供低成本生物标志物，但现有深度模型将频谱视为静态图像，固定通道间图结构，并忽略先验知识，限制了准确性和可解释性。我们提出ELPG-DTFS，一种基于先验引导的自适应时频图神经网络，包含：(1) 带有跨频带互信息的通道-频带注意力，(2) 可学习邻接矩阵以捕捉动态功能连接，(3) 注入神经科学先验的残差知识图路径。在128通道MODMA数据集（53名受试者）上，ELPG-DTFS达到97.63%准确率和97.33% F1，优于2025年最新方法ACM-GNN。消融实验表明，移除任一模块使F1下降最高达4.35，证实其互补价值。该方法为下一代基于EEG的MDD诊断提供了鲁棒且可解释的框架。
原文摘要 · Abstract (English)
Timely and objective screening of major depressive disorder (MDD) is vital, yet diagnosis still relies on subjective scales. Electroencephalography (EEG) provides a low-cost biomarker, but existing deep models treat spectra as static images, fix inter-channel graphs, and ignore prior knowledge, limiting accuracy and interpretability. We propose ELPG-DTFS, a prior-guided adaptive time-frequency graph neural network that introduces: (1) channel-band attention with cross-band mutual information, (2) a learnable adjacency matrix for dynamic functional links, and (3) a residual knowledge-graph pathway injecting neuroscience priors. On the 128-channel MODMA dataset (53 subjects), ELPG-DTFS achieves 97.63% accuracy and 97.33% F1, surpassing the 2025 state-of-the-art ACM-GNN. Ablation shows that removing any module lowers F1 by up to 4.35, confirming their complementary value. ELPG-DTFS thus offers a robust and interpretable framework for next-generation EEG-based MDD diagnostics.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。