Li Tianyi, Park Seo-Hyun, Lee Changwoo, Kim Shawn, Kwon Younghoon, Kim Hojun, Chung Jae-Hyun
Department of Mechanical Engineering, University of Washington, Box 352600, Seattle, WA 98195, USA.
Department of Rehabilitation Medicine of Korean Medicine, Dongguk University, Bundang Hospital, Seongnam, Republic of Korea, 13601; Department of Korean Medicine, Dongguk University WISE campus, Gyeongju, Republic of Korea, 38066.
Adv Sens Res. 2025 Jul;4(7). doi: 10.1002/adsr.202500027. Epub 2025 May 22.
Fatigue negatively impacts health, safety, and productivity, yet current monitoring methods are often subjective, labor-intensive, and inaccurate. To address these challenges, this study presents a capacitive sensor-based eye tracker leveraging cylindrical carbon nanotube-paper composite (CCPC) sensors for chronic fatigue (CF) assessment. Fabricated by novel wet-fracture and paper-rolling methods, CCPC sensors demonstrate superior proximity sensitivity with a small form factor. These one-dimensional sensors are seamlessly integrated into an eyeglass frame for noncontact monitoring of blink rates and eye closures. A 15-minute testing protocol, combining cognitive tasks and noise exposure, is designed to induce acute fatigue and identify CF. By analyzing changes in the digital markers against established fatigue indicators, CF is assessed with the aid of machine learning models for the evaluation of accuracy, sensitivity, and specificity. This real-time, wearable monitoring platform provides an objective, effortless, and noncontact approach to fatigue assessment. With further testing and optimization, it holds the potential for user-friendly evaluation of acute fatigue or fatigue-associated diseases, such as myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).
疲劳会对健康、安全和生产力产生负面影响，但目前的监测方法往往主观、劳动强度大且不准确。为应对这些挑战，本研究提出了一种基于电容式传感器的眼动追踪器，利用圆柱形碳纳米管纸复合材料（CCPC）传感器进行慢性疲劳（CF）评估。通过新颖的湿断裂和卷纸方法制造的CCPC传感器具有外形小巧、接近灵敏度高的优点。这些一维传感器被无缝集成到眼镜框架中，用于非接触式监测眨眼率和闭眼情况。设计了一个15分钟的测试方案，结合认知任务和噪声暴露，以诱发急性疲劳并识别慢性疲劳。通过分析数字标记相对于既定疲劳指标的变化，借助机器学习模型评估准确性、灵敏度和特异性来评估慢性疲劳。这个实时、可穿戴的监测平台为疲劳评估提供了一种客观、轻松且非接触的方法。经过进一步测试和优化，它有望实现对急性疲劳或疲劳相关疾病（如肌痛性脑脊髓炎/慢性疲劳综合征（ME/CFS））的用户友好型评估。
