Abstract
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.
Similar content being viewed by others
References
Anders, S., Lotze, M., Erb, M., Grodd, W., & Birbaumer, N. (2004). Brain activity underlying emotional valence and arousal: a response-related fmri study. Human Brain Mapping, 23(4), 200–209.
Belk, M., Portugal, D., Germanakos, P., Quintas, J., Christodoulou, E., & Samaras, G. (2016). A computer mouse for stress identification of older adults at work. In UMAP (extended proceedings).
Betti, S., Lova, R.M., Rovini, E., Acerbi, G., Santarelli, L., Cabiati, M., Ry, S.D., & Cavallo, F. (2018). Evaluation of an integrated system of wearable physiological sensors for stress monitoring in working environments by using biological markers. IEEE Transactions on Biomedical Engineering, 65 (8), 1748–1758. https://doi.org/10.1109/TBME.2017.2764507.
Birkett, M. (2011). The trier social stress test protocol for inducing psychological stress. Journal of Visualized Experiments: JoVE. https://doi.org/10.3791/3238.
Björntorp, P., & Rosmond, R. (2000). Obesity and cortisol. Nutrition, 16(10), 924–936.
Bombardini, T., Gemignani, V., Bianchini, E., Venneri, L., Petersen, C., Pasanisi, E., Pratali, L., Alonso-Rodriguez, D., Pianelli, M., Faita, F., & et al. (2008). Diastolic time–frequency relation in the stress echo lab: filling timing and flow at different heart rates. Cardiovascular Ultrasound, 6(1), 15.
Bozovic, D., Racic, M., & Ivkovic, N. (2013). Salivary cortisol levels as a biological marker of stress reaction. Medical Archives, 67(5), 374–377.
Buijs, R.M., & Kalsbeek, A. (2001). Hypothalamic integration of central and peripheral clocks. Nature Reviews Neuroscience, 2(7), 521–526.
Charles, S.T. (2010). Strength and vulnerability integration: a model of emotional well-being across adulthood. Psychological Bulletin, 136(6), 1068.
Chrousos, G. (2009). Stress and disorders of the stress system. Nature reviews. Endocrinology, 5, 374–81. https://doi.org/10.1038/nrendo.2009.106.
Critchley, H.D. (2002). Electrodermal responses: what happens in the brain. The Neuroscientist, 8(2), 132–142.
Delmastro, F., Martino, F.D., & Dolciotti, C. (2020). Cognitive training and stress detection in mci frail older people through wearable sensors and machine learning. IEEE Access, 8, 65573–65590.
Dickerson, S., & Kemeny, M. (2004). Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355–91. https://doi.org/10.1037/0033-2909.130.3.355.
Folkman, S., Lazarus, R.S., Pimley, S., & Novacek, J. (1987). Age differences in stress and coping processes. Psychology and Aging, 2(2), 171.
Fulop, T., Larbi, A., Witkowski, J.M., McElhaney, J., Loeb, M., Mitnitski, A., & Pawelec, G. (2010). Aging, frailty and age-related diseases. Biogerontology, 11(5), 547–563.
Gerra, G., Monti, D., Panerai, A.E., Sacerdote, P., Anderlini, R., Avanzini, P., Zaimovic, A., Brambilla, F., & Franceschi, C. (2003). Long-term immune-endocrine effects of bereavement: relationships with anxiety levels and mood. Psychiatry Research, 121(2), 145–158.
Graham, J.E., Christian, L.M., & Kiecolt-Glaser, J.K. (2006). Stress, age, and immune function: toward a lifespan approach. Journal of Behavioral Medicine, 29(4), 389–400.
Greene, S., Thapliyal, H., & Caban-Holt, A. (2016). A survey of affective computing for stress detection: evaluating technologies in stress detection for better health. IEEE Consumer Electronics Magazine, 5 (4), 44–56. https://doi.org/10.1109/MCE.2016.2590178.
Henry, J.D., & Crawford, J.R. (2005). The short-form version of the depression anxiety stress scales (dass-21): construct validity and normative data in a large non-clinical sample. British Journal of Clinical Psychology, 44(2), 227–239.
Juster, R.P., McEwen, B.S., & Lupien, S.J. (2010). Allostatic load biomarkers of chronic stress and impact on health and cognition. Neuroscience & Biobehavioral Reviews, 35(1), 2–16.
Kiecolt-Glaser, J.K., & Glaser, R. (1991). Stress and immune function in humans. In R. Ader, D.L. Felten, & N. Cohen (Eds.) Psychoneuroimmunology. https://doi.org/10.1016/B978-0-12-043780-1.50038-5 (pp. 849–867): Academic Press.
Kikhia, B., Stavropoulos, T.G., Andreadis, S., Karvonen, N., Kompatsiaris, I., Sävenstedt, S., Pijl, M., & Melander, C. (2016). Utilizing a wristband sensor to measure the stress level for people with dementia. Sensors, 16(12), 1989.
Kulkarni, S., O’Farrell, I., Erasi, M., & Kochar, M. (1998). Stress and hypertension. WMJ: Official Publication of the State Medical Society of Wisconsin, 97(11), 34–38.
Liao, W., Zhang, W., Zhu, Z., & Ji, Q. (2005). A real-time human stress monitoring system using dynamic bayesian network. In 2005 IEEE Computer Society conference on computer vision and pattern recognition (CVPR’05)-workshops (pp. 70–70): IEEE.
Lin, W.H., Wu, D., Li, C., Zhang, H., & Zhang, Y.T. (2014). Comparison of heart rate variability from ppg with that from ecg. In The international conference on health informatics (pp. 213–215): Springer.
Luecken, L.J., & Lemery, K.S. (2004). Early caregiving and physiological stress responses. Clinical Psychology Review, 24(2), 171–191.
Marteau, T.M., & Bekker, H. (1992). The development of a six-item short-form of the state scale of the spielberger state—trait anxiety inventory (stai). British Journal of Clinical Psychology, 31(3), 301–306.
Martinez, H.P., Bengio, Y., & Yannakakis, G.N. (2013). Learning deep physiological models of affect. IEEE Computational Intelligence Magazine, 8(2), 20–33. https://doi.org/10.1109/MCI.2013.2247823.
Martini, F., & et al. (2006). Anatomy and Physiology’2007 Ed. Rex Bookstore, Inc.
McEwen, B.S. (1998). Stress, adaptation, and disease: allostasis and allostatic load. Annals of the New York Academy of Sciences, 840(1), 33–44.
McEwen, B.S. (2008). Central effects of stress hormones in health and disease: understanding the protective and damaging effects of stress and stress mediators. European Journal of Pharmacology, 583(2–3), 174–185.
Miller, G., Rohleder, N., Stetler, C., & Kirschbaum, C. (2005). Clinical depression and regulation of the inflammatory response during acute stress. Psychosomatic Medicine, 67(5), 679–687. https://doi.org/10.1097/01.psy.0000174172.82428.ce.
Nath, R.K., Thapliyal, H., Caban-Holt, A., & Mohanty, S.P. (2020). Machine learning based solutions for real-time stress monitoring. IEEE Consumer Electronics Magazine, 9(5), 34–41.
Oswald, L.M., Zandi, P., Nestadt, G., Potash, J.B., Kalaydjian, A.E., & Wand, G. S. (2006). Relationship between cortisol responses to stress and personality. Neuropsychopharmacology, 31(7), 1583–1591.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, E. (2011). Scikit-learn: machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
Rothman, S.M., & Mattson, M.P. (2010). Adverse stress, hippocampal networks, and alzheimer’s disease. Neuromolecular Medicine, 12(1), 56–70.
Seeman, T.E., McEwen, B.S., Singer, B.H., Albert, M.S., & Rowe, J.W. (1997). Increase in urinary cortisol excretion and memory declines: Macarthur studies of successful aging. The Journal of Clinical Endocrinology & Metabolism, 82(8), 2458–2465.
Segerstrom, S.C., & Miller, G.E. (2004). Psychological stress and the human immune system: a meta-analytic study of 30 years of inquiry. Psychological Bulletin, 130(4), 601.
Seo, W., Kim, N., Kim, S., Lee, C., & Park, S.M. (2019). Deep ecg-respiration network (deeper net) for recognizing mental stress. Sensors, 19, 3021. https://doi.org/10.3390/s19133021.
Setz, C., Arnrich, B., Schumm, J., La Marca, R., Tröster, G., & Ehlert, U. (2010). Discriminating stress from cognitive load using a wearable eda device. IEEE Transactions on Information Technology in Biomedicine, 14(2), 410–417.
Shaughnessy, J.J., Zechmeister, E.B., & Zechmeister, J.S. (2000). Research methods in psychology. New York: McGraw-Hill.
Smets, E., De Raedt, W., & Van Hoof, C. (2019). Into the wild: the challenges of physiological stress detection in laboratory and ambulatory settings. IEEE Journal of Biomedical and Health Informatics, 23(2), 463–473.
Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information Processing & Management, 45(4), 427–437.
Thapliyal, H., Khalus, V., & Labrado, C. (2017). Stress detection and management: a survey of wearable smart health devices. IEEE Consumer Electronics Magazine, 6(4), 64–69.
Acknowledgments
This work was supported by the Kentucky Science and Engineering Foundation under Grant KSEF-3528-RDE-019.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Background on LSTM
Appendix: Background on LSTM
A single LSTM cell consists of three inputs: (i) Input at current time step, xt, (ii) Hidden state at previous time step, ht− 1 and (iii) Cell state at the previous time step, ct− 1. These inputs undergo a series of mathematical operations to give the cell state at the current time step, ct and the hidden state of the current time step, ht.
The cell state and hidden state at present time is estimated using the following steps:
-
First ht− 1 and xt is combined and passed through a sigmoid function. This is also known as forget gate. Sigmoid function bounds the output in between 0 and 1. A value of 0 means that particular information needs to be forgotten while values close to 1 indicates the high importance of the information. The output of the forget gate (ft) is then subjected to a point wise multiplication with ct− 1.
-
The output (It) of the sigmoid operation on xt + ht− 1 is multiplied point wise with the output (\(c^{\prime }_{t}\)) to generate the expression \(I_{t}+c^{\prime }_{t}\).
-
The present cell state is then given by the expression \(c_{t}=c_{t-1}*f_{t}+I_{t}*c^{\prime }_{t}\)
-
ht is calculated by performing point wise multiplication on the output from the sigmoid operation on xt + ht− 1 and ct.
Rights and permissions
About this article
Cite this article
Nath, R.K., Thapliyal, H. & Caban-Holt, A. Machine Learning Based Stress Monitoring in Older Adults Using Wearable Sensors and Cortisol as Stress Biomarker. J Sign Process Syst 94, 513–525 (2022). https://doi.org/10.1007/s11265-020-01611-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11265-020-01611-5