跳到正文
arXiv cs.AI· Gurinder Kaur Sandhu, Jude Koenig, Ankita Shukla, Shruti Kshirsagar·· 4 小时前AI 评分31

基于「SleepXAI」单通道脑电的可解释睡眠分期研究

Towards Explainable Sleep Stage Classification using Single-Channel Electroencephalogram

AI 导读

研究者扩展 SleepXAI 睡眠分期框架,系统比较五种 XAI 方法,聚焦更易临床部署的单通道 EEG。在 Sleep-EDF 与 DREAMT 数据集上,以 Spearman 秩相关对照 AASM 定义的 EEG 频段评估归因保真度,其中 Integrated Gradients(IG)在 Sleep-EDF 全部五个睡眠分期及 DREAMT 各 AHI 分组中均取得最高平均归因保真度。

正文 · 原文

View PDF HTML (experimental)

Abstract:Automated sleep staging has achieved strong bench- mark accuracy; however, clinical adoption remains limited due to a lack of interpretability. In this work, we fill the research gap by extending the SleepXAI sleep staging framework to systematically compare five different XAI methods. Further, attribution fidelity is evaluated using Spearman rank correlation with AASM-defined EEG frequency bands, serving as clinically grounded reference signals. In this paper, we focus on single- channel Electroencephalogram (EEG), which is more amenable to practical clinical deployment compared to full polysomnography setups. Experiments are conducted on the Sleep-EDF benchmark dataset and the DREAMT dataset. In DREAMT, we stratified subjects based on OSA into four Apnea-Hypopnea Index (AHI) severity groups. We observed that the SleepXAI model achieves strong classification performance on both datasets. Among the evaluated methods, Integrated Gradients (IG) consistently achieves the highest mean attribution fidelity across all five sleep stages on Sleep-EDF and across all AHI groups on DREAMT, outperforming all other evaluated methods under the Spearman rank correlation metric. These findings highlight the importance of attribution method selection as a key factor for ensuring reliable and clinically interpretable automated sleep staging systems.
Comments: 8 pages, 4 figures. Accepted at AgIA4BioHealth Workshop, IEEE BIBM 2026, Dallas, Texas, December 1-4, 2026
Subjects: Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
Cite as: arXiv:2611.18434 [cs.AI]
  (or arXiv:2611.18434v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.18434

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Gurinder Kaur Sandhu [view email]
[v1] Fri, 13 Nov 2026 17:13:23 UTC (1,028 KB)

来源:arXiv cs.AI · arxiv.org