UMAP Embeddings of Sleep-EDF-78

<h1><span dir="auto"><span dir="auto">AISleep: Automated and interpretable sleep staging from single-channel EEG data<br><br></span></span></h1> <p><span lang="EN-US">Xun Mai</span><span lang="EN-US">,</span><span lang="EN-US"> Binghua Song, Manli Luo</span><span lang="EN-US">, Jun Zhu, Xu Jiang, Xiao Ma</span><span lang="EN-US">, Feng Lin</span><span lang="EN-US">, Xiaoqing Hu⁵, Hanchuan Peng^*, Li Zhang^*, </span><span lang="EN-US">Y</span><span lang="EN-US">ina Wei^*</span></p> <h2><span lang="EN-US">^<br>Abstract</span></h2> <p><span lang="EN-US">^{<br>Sleep staging is essential for understanding sleep physiology and diagnosing sleep-related disorders. However, traditional manual scoring is time-consuming and resource-intensive, limiting its scalability for large-scale application. In this study, we introduce AISleep, an automated and interpretable unsupervised algorithm based on feature-weighted kernel density estimation (KDE), designed to stage sleep using only a single electroencephalography (EEG) channel. AISleep was evaluated using both public benchmark datasets of healthy subjects and clinical datasets from patients with sleep disorder. It outperforms state-of-the-art (SOTA) unsupervised sleep staging algorithms in young, healthy subjects, and demonstrates better generalizability compared to supervised models. Importantly, we observed that some key EEG features decline with age, which may contribute to reduced staging accuracy in older adults. This study presents a robust and interpretable unsupervised sleep staging algorithm whose lightweight design makes it well-suited for integration into portable devices, offering a practical and scalable solution for accurate, home-based sleep monitoring.<br>}</span></p> <p>&nbsp;</p> <p><span dir="auto"><span dir="auto">&nbsp;</span></span></p> <p>&nbsp;</p>