Experimental Study of the Stress Level at the Workplace Using an Smart Testbed of Wireless Sensor Networks and Ambient Intelligence Techniques

Abstract

Abstract. This paper combines techniques of ambient intelligence and wireless sensor networks with the objective of obtain important conclusions to increase the quality of life of people. In particular, we oriented our study to the stress at the workplace, because stress is a leading cause of illness and disease. This article presents a wireless sensor network obtaining information of the environment, a pulse sensor obtaining hear rate values and a complete data analysis applying techniques of ambient intelligence to predict stress from these environment variables and people attributes. Results show promise on the identification of stressful situations as well as stress inference through the use of predictive algorithms.

Context-Aware Well-Being Assessment in Intelligent Environments

“…This envision may also help annotate different states of comfort through building and the people inside to better map how the environmental attributes of such buildings affect the well-being of people. As tried in the experimental framework to detect stressful moods (Silva et al 2013), where a testbed was used to perceive different room states during the day and the classification of probable well-being.…”

“…There are however ergonomic studies that provide the necessary background to create an environment which satisfies the most common needs to keep both the people and environments healthy. Although not being an extensive review of possible attributes it does indicate a minimum set of possible attributes to start comfort and well-being assessments [14].…”

Data Mining for Wearable Sensors in Health Monitoring Systems: A Review of Recent Trends and Challenges

“…This approach is also known as supervised learning models [ 20 ] where it includes feature extraction, training and testing steps while performing the prediction of the data behavior. As the common examples of the predictive models, authors in [ 38 , 39 ] presented a method which predicts the further stress levels of a subject. A similar example of using predictive models in healthcare are: blood glucose level prediction [ 23 ], mortality prediction by clustering electronic health data [ 40 ], and a predictive decision making system for dialysis patients [ 41 ].…”

“…In the most of research in health monitoring systems, dealing with the simple statistical parameters (e.g., mean, variance etc. ) or statistical functions e.g., risk function [ 61 ], factor analysis [ 68 ]) would be adequate to formulate the data and retrieve the expected information [ 38 ]. This kind of data analysis more or less applied on multi sensor networks in order to simplify the data features for model construction [ 30 ].…”

“… Experimental wearable sensor data: The papers which have developed the health monitoring systems have mostly used their own data gathering experiments to design, model and test the data analysis step [ 26 , 38 , 88 ]. In this case the gathered data are usually obtained based on the predefined scenarios due to the test and evaluate the performed results [ 36 ], but usually these studies do not provide the precise annotations and meaningful labels on physiological signals.…”