Machine Learning Based Stress Monitoring in Older Adults Using Wearable Sensors and Cortisol as Stress Biomarker

The objective of this work is to evaluate the effectiveness of a wearable physiological stress monitoring system in distinguishing between stressed and non-stressed state in older adults using machine learning techniques. This system utilizes EDA and BVP signal to detect occurrence of stress as indicated by salivary cortisol measurement which is a reliable objective measure of physiological stress. Data of 19 healthy older adults (11 female and 8 male) with mean age 73.15 ± 5.79 were used for this study. EDA and BVP signals were recorded using a finger tip sensor during the Trier Social Stress Test, which is a well known experimental protocol to reliably induce stress in humans in a social setting. 39 statistical measures of the peak characteristic of EDA and BVP signal were extracted. A supervised feature selection algorithm is used to select important features as an input to the machine learning model. Four machine learning algorithms were evaluated based on their performance in classifying between stressed and non-stressed states. Results indicate that the logistic regression performed the best among Random Forest, κ-NN, and Support Vector Machine achieving an macro-average and micro-average f1-score of 0.87 and 0.95 respectively and an AUC score of 0.81. We also evaluated the effectiveness of a novel deep learning Long Short-Term Memory (LSTM) based classifier in distinguishing between stressed and non-stressed state. Results on test data shows that LSTM based classifier achieved an improvement of 6.7% and 2% in terms of macro-average f1-score and micro-average f1-score respectively. Also the AUC score for LSTM classifier is found to be 0.9 which is about 11% higher than the best performing logistic regression model. This work can be used to design a convenient unobtrusive wearable device to monitor stress levels in older adults in their home environment, thereby facilitating aging in place and improving the quality of life.

[1]  Himanshu Thapliyal,et al.  A Survey of Affective Computing for Stress Detection: Evaluating technologies in stress detection for better health , 2016, IEEE Consumer Electronics Magazine.

[2]  M. Mattson,et al.  Adverse Stress, Hippocampal Networks, and Alzheimer’s Disease , 2010, NeuroMolecular Medicine.

[3]  E. B. Zechmeister,et al.  Research Methods in Psychology. , 1990 .

[4]  S. Folkman,et al.  Age differences in stress and coping processes. , 1987, Psychology and aging.

[5]  M. Loeb,et al.  Aging, frailty and age-related diseases , 2010, Biogerontology.

[6]  Filippo Cavallo,et al.  Evaluation of an Integrated System of Wearable Physiological Sensors for Stress Monitoring in Working Environments by Using Biological Markers , 2018, IEEE Transactions on Biomedical Engineering.

[7]  M. Kochar,et al.  Stress and hypertension. , 1998, WMJ : official publication of the State Medical Society of Wisconsin.

[8]  B. McEwen.  Central effects of stress hormones in health and disease: Understanding the protective and damaging effects of stress and stress mediators. , 2008, European journal of pharmacology.

[9]  Zhiwei Zhu,et al.  A Real-Time Human Stress Monitoring System Using Dynamic Bayesian Network , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[10]  Chanhee Lee,et al.  Deep ECG-Respiration Network (DeepER Net) for Recognizing Mental Stress , 2019, Sensors.

[11]  Basel Kikhia,et al.  Utilizing a Wristband Sensor to Measure the Stress Level for People with Dementia , 2016, Sensors.

[12]  J. Henry,et al.  The short-form version of the Depression Anxiety Stress Scales (DASS-21): construct validity and normative data in a large non-clinical sample. , 2005, The British journal of clinical psychology.

[13]  Gerhard Tröster,et al.  Discriminating Stress From Cognitive Load Using a Wearable EDA Device , 2010, IEEE Transactions on Information Technology in Biomedicine.

[14]  Yoshua Bengio,et al.  Learning deep physiological models of affect , 2013, IEEE Computational Intelligence Magazine.

[15]  A. Kalsbeek,et al.  Hypothalamic integration of central and peripheral clocks , 2001, Nature Reviews Neuroscience.

[16]  Carson Labrado,et al.  Stress Detection and Management: A Survey of Wearable Smart Health Devices , 2017, IEEE Consumer Electronics Magazine.

[17]  M. Birkett.  The Trier Social Stress Test protocol for inducing psychological stress. , 2011, Journal of visualized experiments : JoVE.

[18]  Maja Racic,et al.  Salivary cortisol levels as a biological marker of stress reaction. , 2013, Medical archives.

[19]  Panagiotis Germanakos,et al.  A Computer Mouse for Stress Identification of Older Adults at Work , 2016, UMAP.

[20]  G. Chrousos.  Stress and disorders of the stress system , 2009, Nature Reviews Endocrinology.

[21]  Bruce Mcewen,et al.  Stress, Adaptation, and Disease: Allostasis and Allostatic Load , 1998, Annals of the New York Academy of Sciences.

[22]  B. McEwen,et al.  Allostatic load biomarkers of chronic stress and impact on health and cognition , 2010, Neuroscience & Biobehavioral Reviews.

[23]  Jennifer E. Graham,et al.  Stress, Age, and Immune Function: Toward a Lifespan Approach , 2006, Journal of Behavioral Medicine.

[24]  Guy Lapalme,et al.  A systematic analysis of performance measures for classification tasks , 2009, Inf. Process. Manag..

[25]  S. Dickerson,et al.  Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. , 2004, Psychological bulletin.

[26]  T. Marteau,et al.  The development of a six-item short-form of the state scale of the Spielberger State-Trait Anxiety Inventory (STAI). , 1992, The British journal of clinical psychology.

[27]  Saraju P. Mohanty,et al.  Machine Learning Based Solutions for Real-Time Stress Monitoring , 2020, IEEE Consumer Electronics Magazine.

[28]  J. Kiecolt-Glaser,et al.  Stress and Immune Function in Humans , 1991 .

[29]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[30]  P. Björntorp,et al.  Obesity and cortisol. , 2000, Nutrition.

[31]  Franca Delmastro,et al.  Cognitive Training and Stress Detection in MCI Frail Older People Through Wearable Sensors and Machine Learning , 2020, IEEE Access.

[32]  C. Kirschbaum,et al.  Clinical Depression and Regulation of the Inflammatory Response During Acute Stress , 2005, Psychosomatic medicine.

[33]  S. Charles.  Strength and vulnerability integration: a model of emotional well-being across adulthood. , 2010, Psychological bulletin.

[34]  L. Luecken,et al.  Early caregiving and physiological stress responses. , 2004, Clinical psychology review.

[35]  Francesco Faita,et al.  Diastolic time – frequency relation in the stress echo lab: filling timing and flow at different heart rates , 2008, Cardiovascular ultrasound.

[36]  Gary S Wand,et al.  Relationship between Cortisol Responses to Stress and Personality , 2006, Neuropsychopharmacology.

[37]  Elena Smets,et al.  Into the Wild: The Challenges of Physiological Stress Detection in Laboratory and Ambulatory Settings , 2019, IEEE Journal of Biomedical and Health Informatics.

[38]  H. Critchley.  Review: Electrodermal Responses: What Happens in the Brain , 2002 .

[39]  Wan-Hua Lin,et al.  Comparison of Heart Rate Variability from PPG with That from ECG , 2014 .

[40]  C. Franceschi,et al.  Long-term immune-endocrine effects of bereavement: relationships with anxiety levels and mood , 2003, Psychiatry Research.

[41]  M. Erb,et al.  Brain activity underlying emotional valence and arousal: A response‐related fMRI study , 2004, Human brain mapping.

[42]  M. Albert,et al.  Journal of Clinical Endocrinology and Metabolism Printed in U.S.A. Copyright © 1997 by The Endocrine Society Increase in Urinary Cortisol Excretion and Memory Declines: MacArthur Studies of Successful Aging* , 2022 .

[43]  S. Segerstrom,et al.  Psychological stress and the human immune system: a meta-analytic study of 30 years of inquiry. , 2004, Psychological bulletin.