Real-Time Personalized Physiologically Based Stress Detection For Hazardous Operations
Keywords:
Stress monitoring, AI-driven analysis, biometric sensors, immersive simulations, astronaut training.Abstract
This study explores a real-time stress detection system designed for hazardous operations, aiming to enhance performance and reduce stress. Traditional machine-learning models struggle with stress detection due to individual differences and the time-series nature of physiological signals. To address this, a personalized model was developed, selecting specific features for training. The system was tested for real-time deployment using physiological data—heart rate, blood pressure, electrodermal activity, and respiration—collected from participants performing tasks with varying stress levels.
A comparison of classifiers, including Support Vector Machine, Decision Tree, Random Forest, and an Approximate Bayes (A Bayes) classifier, showed that personalized models outperform generalized ones in classifying stress levels. Results indicate that model accuracy varies with feature selection, window size, and task type, with blood pressure emerging as a key indicator. The study highlights the advantage of personalized models in stress detection and their potential for future applications.
Downloads
References
J. E. Driskell, E. Salas, J. H. Johnston, and T. N. Wollert, Stress Exposure Training: An Event-Based Approach (Performance Under Stress). London, U.K.: Ashgate, 2008, pp. 271–286.
[2] I. Barshi and D. L. Dempsey, ‘‘Risk of performance errors due to training deficiencies: Evidence report,’’ Nat. Aeronaut.
Space Admin. (NASA), NASA Johnson Space Center, Houston, TX, USA, Tech. Rep., JSCCN- 35755, 2016.
[3] M. Gjoreski, M. Luštrek, M. Gams, and H. Gjoreski,
‘‘Monitoring stress with a wrist device using context,’’ J. Biomed. Inform., vol. 73, pp. 159–170, Sep. 2017, doi: 10.1016/j.jbi.2017.08.006.
[4] M. Zahabi and A. M. A. Razak, ‘‘ Adaptive virtual reality- grounded training A methodical literature review and frame,’’ Virtual Reality, vol. 24, no. 4, pp. 725 – 752, Dec. 2020, doi 10.1007/ s10055- 020-00434-w.
[5] Y. S. Can, B. Arnrich, and C. Ersoy, ‘‘Stress detection in daily life scenarios using smart phones and wearable sensors: A survey,’’ J. Biomed. Informat., vol. 92, Apr. 2019, Art. no.103139, doi:10.1016/j.jbi.2019.103139.
[6] Akmandor, A. O., & Jha, N. K.( 2017)." Keep the stress down with Soda pop A stress discovery and relief system." IEEE Deals onMulti-Scale Computing Systems, 3( 4), 269 – 282.
[7] S. Tong and D. Koller, ‘‘Bayes optimal hyperplanes? Maximal margin hyperplanes,’’ in Proc. IJCAI, 1999, pp. 1–5.
[8] I. Rish, ‘‘An empirical study of the naive Bayes classifier,’’ in
Proc. IJCAI Workshop Empirical Methods Artif. Intell., 2001, vol. 3, no. 22, pp. 41–46.
[9] F. Shaffer, R. McCraty, and C. L. Zerr, ‘‘ A healthy heart is n't a metronome An integrative review of the heart’s deconstruction and heart rate variability,’’ borders Psychol., vol. 5, p. 1040, Sep. 2014, doi 10.3389/ fpsyg.2014.01040.
[10] B. Kim, Y.- S. Jeong, and M. K. Jeong, ‘‘ New multivariate kernel viscosity estimator for uncertain data bracket,’’ Ann. Oper. Res., vol. 303, nos. 1 – 2, pp. 413 – 431, Aug. 2021.
[11] E. Smets, W. De Raedt, and C. Van Hoof, ‘‘ Into the wild The challenges of physiological stress discovery in laboratory and itinerant settings,’’ IEEE J. Biomed. Health Informat., vol. 23, no. 2, pp. 463 – 473, Mar. 2019, doi 10.1109/ JBHI.2018.2883751.
[12] D. Jones and S. Dechmerowski, ‘‘Measuring stress in an augmented
