IEEE J Biomed Health Inform. 2019 Nov;23(6):2276-2285. doi: 10.1109/JBHI.2019.2938311. Epub 2019 Aug 29.
Mental illnesses are vast and cause a lot of individual and social discomfort, with significant healthcare costs associated in terms of diagnosis and treatment. They can be triggered by a number of factors including stress, fatigue or anxiety. The associated physiological, cardiac and autonomic changes can be assessed, centrally, through brain imaging or, peripherally, by other signal recording modalities. With recent advances in wearable devices, these parameters can now be assessed in natural living conditions as associated mood disorders such as obsessive/compulsive behavior or depression are difficult to be examined in controlled settings. In this paper, we present a low-powered and flexible device with electrocardiogram (ECG), galvanic skin response (GSR), temperature and bio-motion detection channels, with signal accuracies of 62 μV for ECG, 6.6 kΩ for GSR, 0.13 °C for temperature and 0.04 g for acceleration. Potential applications include mental health assessment of patients during daily activities at home and/or work through non-continuous and multimodal sensing as demonstrated in this paper during exercise, rest and mental activities performed by healthy individuals only, achieving an overall accuracy of 89% in the classification of the different tasks executed by volunteers.
精神疾病种类繁多,给个人和社会带来了很多不适,在诊断和治疗方面也带来了巨大的医疗成本。它们可能由多种因素引发,包括压力、疲劳或焦虑。相关的生理、心脏和自主变化可以通过脑成像进行中心评估,也可以通过其他信号记录模式进行外周评估。随着可穿戴设备的最新进展,这些参数现在可以在自然生活条件下进行评估,因为与情绪障碍相关的行为,如强迫症或抑郁症,在受控环境中很难进行检查。在本文中,我们提出了一种低功耗、灵活的设备,具有心电图(ECG)、皮肤电反应(GSR)、温度和生物运动检测通道,ECG 的信号精度为 62 μV,GSR 为 6.6 kΩ,温度为 0.13°C,加速度为 0.04 g。潜在的应用包括通过非连续和多模态传感对在家中和/或工作中的患者进行心理健康评估,如本文所示,在健康个体进行运动、休息和心理活动期间进行,志愿者执行的不同任务的分类准确率达到 89%。