"Stress overload” (SO) represents an imbalance where psychosocial, work, environmental, or health-related burdens exceed an individual's adaptive capacity, engendering negative physical and mental health outcomes. This study reviews tech tools, focusing on Heart Rate Variability (HRV) and metrics for sleep quality analysis (SA) i.e. the number of nocturnal awakenings (NNA). Modern technologies for measuring HRV and NNA could enhance the traditional stress detection methods, respectively from the neuropsychologist by the “Stress Overload Scale” (SOS) and polysomnography (PSG) by the neurophysiology technician. Notably, they also allow continuous monitoring, and may curtail healthcare costs. Employing Machine Learning (ML) on smartwatches using photoplethysmography (PPG) HRV accuracy achieves 98.10-98.18%. Wearable devices also exhibit strong sensitivity and specificity for measuring NNA. In children distinct nocturnal movement patterns should be considered. While HRV correlates directly with stress levels, poor SA indicates only a 4.7 fold increased risk of SO. HRV's integration into allostatic load assessments is advocated. However clinical validation is necessary, while potential privacy concerns may arise, as electrocardiography (ECG) signals can potentially uniquely identify patients. Synthesizing HRV solely from photoplethysmography (PPG) data obtained from wearable devices offers an economical and practical approach, although it may be less accurate than HRV from ECG guided by Respiratory Rate (RR). The latter requires additional hardware, i.e. ECG sensor and either Respiratory Inductive Plethysmography or Respiratory Impedance Plethysmography. However, modern smartwatches possess sufficient computational power to perform ML inference, making them capable of improving PPG-based HRV estimation. By leveraging comprehensive datasets capturing signals like plethysmography, ECG, HRV, and RR, alongside sleep metrics, it will be possible to develop refined algorithms to increase the accuracy of stress risk prediction and its level. Continuous monitoring may support nursing diagnosis of SO, enhancing early intervention and its evaluation. © 2023 CEUR-WS. All rights reserved.
Assessment of Stress Levels using technological tools: A Review and Prospective Analysis of Heart Rate Variability and Sleep Quality Parameters
Comparcini D.;Napolitano D.;Cicolini G.
2023-01-01
Abstract
"Stress overload” (SO) represents an imbalance where psychosocial, work, environmental, or health-related burdens exceed an individual's adaptive capacity, engendering negative physical and mental health outcomes. This study reviews tech tools, focusing on Heart Rate Variability (HRV) and metrics for sleep quality analysis (SA) i.e. the number of nocturnal awakenings (NNA). Modern technologies for measuring HRV and NNA could enhance the traditional stress detection methods, respectively from the neuropsychologist by the “Stress Overload Scale” (SOS) and polysomnography (PSG) by the neurophysiology technician. Notably, they also allow continuous monitoring, and may curtail healthcare costs. Employing Machine Learning (ML) on smartwatches using photoplethysmography (PPG) HRV accuracy achieves 98.10-98.18%. Wearable devices also exhibit strong sensitivity and specificity for measuring NNA. In children distinct nocturnal movement patterns should be considered. While HRV correlates directly with stress levels, poor SA indicates only a 4.7 fold increased risk of SO. HRV's integration into allostatic load assessments is advocated. However clinical validation is necessary, while potential privacy concerns may arise, as electrocardiography (ECG) signals can potentially uniquely identify patients. Synthesizing HRV solely from photoplethysmography (PPG) data obtained from wearable devices offers an economical and practical approach, although it may be less accurate than HRV from ECG guided by Respiratory Rate (RR). The latter requires additional hardware, i.e. ECG sensor and either Respiratory Inductive Plethysmography or Respiratory Impedance Plethysmography. However, modern smartwatches possess sufficient computational power to perform ML inference, making them capable of improving PPG-based HRV estimation. By leveraging comprehensive datasets capturing signals like plethysmography, ECG, HRV, and RR, alongside sleep metrics, it will be possible to develop refined algorithms to increase the accuracy of stress risk prediction and its level. Continuous monitoring may support nursing diagnosis of SO, enhancing early intervention and its evaluation. © 2023 CEUR-WS. All rights reserved.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
