发表机构
School of Mathematics and Systems Science, Shandong University of Science and Technology; School of Artificial Intelligence, Shandong Women’s University（山东科技大学数学与系统科学学院; 山东女子学院人工智能学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有睡眠分期方法在模糊及过渡分期表现差的问题，提出分层上下文感知的双流框架 SwinSleepNet，在 Sleep-EDF 系列及 SHHS 数据集上验证其在 N1 分期和过渡 epoch 上的性能优势。
AI 中文摘要
自动睡眠分期在睡眠障碍诊断、睡眠质量评估和长期健康监测中发挥关键作用，但现有方法在模糊及过渡相关睡眠分期上表现不佳，原因在于对细粒度的 epoch 内结构和复杂跨区域频谱依赖的建模不足。传统的 epoch 级编码器通常无法提取微妙的时间微观结构和 epoch 内跨区域交互，导致 N1 分期等困难类别的识别准确率不尽如人意。为解决这些缺陷，我们提出 SwinSleepNet，这是一个分层上下文感知的双流框架，可分别优化 epoch 内表示学习和 epoch 间上下文建模。具体而言，我们从两个互补角度表征每个睡眠 epoch：原始时域 EEG 信号及其时频变换。时域分支采用卷积编码器捕捉精细的波形时间细节，时频分支使用 Swin Transformer 提取局部时频谱特征、分层多尺度信息和长程空间依赖。多分支提取的特征被融合为集成嵌入，经双向上下文模块优化以捕捉跨 epoch 的时间依赖，用于最终的睡眠分期分类。在 Sleep-EDF-20、Sleep-EDF-78 和 SHHS 数据集上的综合实验验证，我们的方法实现了有竞争力的整体性能，且在困难的 N1 分期和过渡 epoch 上表现出更强的鲁棒性和稳定性。结果证明，基于分层架构优化的 epoch 内表示学习极大地有益于自动睡眠分期任务。
英文摘要
Automatic sleep staging is a critical role in sleep disorder diagnosis, sleep quality assessment, and long-term health monitoring; however, existing approaches suffer poor performance on ambiguous and transition-related sleep stages, caused by inadequate modeling of fine-grained intra-epoch structures and complex cross-region spectral dependencies. Traditional epoch-level encoders commonly fail to extract subtle temporal microstructures and intra-epoch cross-region interactions, resulting in unsatisfactory recognition accuracy for hard categories such as the N1 stage. To tackle these drawbacks, we propose SwinSleepNet, a hierarchical context-aware dual-stream framework that separately optimizes intra-epoch representation learning and inter-epoch contextual modeling. Concretely, we characterize each sleep epoch from two complementary perspectives: raw time-domain EEG signal and its time-frequency transformation. The time-domain branch adopts convolutional encoders to capture fine waveform temporal details, and the time-frequency branch uses Swin Transformer to extract local spectro-temporal features, hierarchical multi-scale information and long-range spatial dependencies. The multi-branch extracted features are fused into integrated embeddings, which are optimized by a bidirectional context module to capture cross-epoch temporal dependencies for final sleep stage classification. Comprehensive experiments on Sleep-EDF-20, Sleep-EDF-78 and SHHS datasets verify that our method achieves competitive overall performance, and exhibits stronger robustness and stability on difficult N1 stages and transitional epochs. The results prove that optimized intra-epoch representation learning based on hierarchical architecture greatly benefits automatic sleep staging tasks.
CommentsReport-no: SDUST-SLEEP-202608-V2; 10 pages, 7 figures, revised updated version of arXiv submit/7867870, conference submission draft
