The efficacy of transcranial direct current stimulation effect on vigilance decrement predicted by a fuzzy logic model

经颅直流电刺激技术干预警觉度下降效应的警觉度量化方法

¹ School of Electronics and Information, Northwestern Polytechnical University, Xi'an 710072, China
² School of Life Sciences, Northwestern Polytechnical University, Xi'an 710072, China
³ Institute of Medical Research, Northwestern Polytechnical University, Xi'an 710072, China

Received: 23 March 2024

Abstract

Vigilance quantification and fatigue modulation are concerned. However, previous vigilance quantification methods often face limitations in label subjectivity and interpretability when monitoring prolonged task performance. To address these challenges, we established a novel three-layer fuzzy logic framework to explore the feasibility of utilizing the three types of EEG features (amplitudes and latencies of ERPs, power ratio) to predict the vigilance decrement, and to determine a clear mapping relationship for vigilance quantification. Using objective time-on-task behavioral performance, we achieved the classification between vigilance, moderate vigilance, and fatigue, as well as the assessment of intervention effect of transcranial direct current stimulation technology (tDCS) on the vigilance decrement. The results demonstrate that we achieved an average classification accuracy (90.91%) between vigilance, moderate vigilance, and fatigue states, and the tDCS can remediate the vigilance decrement effect to a certain extent.

摘要

准确量化和监测人员在执行长时间任务后的低警觉状态, 并及时进行调控干预, 是目前的研究热点之一。针对当前警觉度监测及警觉度量化方法研究中选取分类标签的主观性强、现有基于深度学习的分类模型可解释性不佳等问题, 提出了一种基于3层模糊逻辑的警觉度量化方法。通过对多个与警觉度下降相关的脑电特征的合并推理, 确定出明确的警觉度量化映射关系, 以长时间任务诱发的客观行为表现作为标签, 实现高警觉、中等警觉、疲劳的三分类, 并评估经颅直流电刺激技术对于警觉度下降效应的干预效果。实验结果证明基于三层模糊逻辑的警觉度量化方法的平均准确度达到90.9%, 在明确了警觉度量化的映射关系的基础上, 验证了经颅直流电刺激在一定程度上能够减弱警觉度下降效应。