基于脉搏波的警觉度检测研究

发布时间：2018-07-17 02:36

【摘要】：本文采用脉搏波信号研究了警觉度的变化规律。本研究共招募10名受试者参加了持续95 min的警觉度"时钟测试"(MCT)。试验过程中,受试者们通过卡罗林斯卡嗜睡量表(KSS)和斯坦福嗜睡量表(SSS)主观评价了清醒程度,同时记录了所有受试者的目标反应时间、目标识别正确率和脉搏波信号。结果表明,根据主观量表得分和受试者的行为学数据可以将警觉度定标为3个水平:前30 min为高警觉度水平;中间30 min为一般警觉度水平,后30 min为低警觉度水平。此外,脉搏波信号的时域特征,如:次级波峰幅值、波峰幅值、次级波峰潜伏期,随警觉度的降低而减小,而波谷幅值随警觉度的降低而增大;频域特征:8.600~9.375 Hz、11.720~12.500 Hz、38.280~39.060 Hz和39.060~39.840 Hz这4个子频带的能量概率也随警觉度的降低而减小。最后,在上述8个特征建立的模型中,10名受试者三分类正确率的平均值高达88.7%。本文的研究结果证实了脉搏波在警觉度评估上的可行性,为警觉度的实时监测提供了新的思路。
[Abstract]:In this paper, the variation of alertness is studied by pulse wave signal. In this study, 10 subjects were recruited to participate in a 95 min alert "clock test" (MCT). In the course of the experiment, the subjects evaluated their waking degree subjectively by the Karolinska sleepiness scale (KSS) and the Stanford somnolence scale (SSS), and recorded the target reaction time, target recognition accuracy and pulse wave signal of all the subjects. The results showed that according to subjective scale scores and behavioral data of subjects, alertness could be calibrated to three levels: the first 30 min was high alert, the middle 30 min was general alertness, and the latter 30 min was low alert. In addition, the time domain characteristics of pulse wave signal, such as the amplitude of the secondary wave peak, the latency of the secondary wave peak, decrease with the decrease of the alertness, while the amplitude of the trough increases with the decrease of the alertness. The energy probability of the four subbands of: 8.600 ~ 9.375 Hzn 11.720 ~ 12.500 Hz ~ 38.280 ~ 39.060 Hz and 39.060 ~ 39.840 Hz also decreases with the decrease of alertness. Finally, the average three-classification accuracy rate of 10 subjects was 88.7in the model established by the above eight features. The results of this paper confirm the feasibility of pulse wave in alertness evaluation and provide a new idea for real-time monitoring of alertness.
【作者单位】：中国航天员科研训练中心人因工程国防科技重点实验室;
【基金】：国家自然科学基金(81671861)
【分类号】：R318

【相似文献】