SOURCE RECORD / S0158
利用基于 EEG 的神经生理频谱特征检测驾驶困倦
Driving drowsiness detection using spectral signatures of EEG-based neurophysiology
- 作者或机构
- Arif et al.
- 发布日期
- 2023
- 来源类型
- peer_reviewed_human_study
- 技术路线
- noninvasive EEG
- 功能场景
- drowsiness detection
- 权利状态
- manual_review / CC BY (version unspecified)
证据摘要
证据支持：在合并数据内部进行困倦分类。关键数字：6 个 EEG channels；合并 10-second 窗口；随机 10-fold；在合并参与者上选择特征。局限：未按参与者或会话分组；存在身份、相邻窗口和全局选择泄漏；男性模拟驾驶样本不能支持跨人群真实场景外推。
关键数字
6 EEG channels; merged 10-second windows; random 10-fold; feature selection on pooled participants
局限
No participant or session grouping; identity adjacent-window and global-selection leakage; male simulated-driving sample cannot support cross-person real-world generalization