Heart Rate Variability Based LSTM Model for Stress Detection with Explainable AI Insights
Keywords:
Explainable -AI, HRV, LSTM, Stress Detection, SWELL DatasetAbstract
In today's busy world, stress is common because people must think about many things simultaneously. To effectively deal with the harmful effects of worry on your health, a person needs to notice them as soon as they appear. This study supports recognizing stress as a helpful method. It shows how critical physiological signs are as a reliable way to detect stress, mainly because these signals cannot be changed purposefully. Heart Rate Variability (HRV), a physiological signal, is used in this study to investigate how stress can be detected using the SWELL knowledge work (SWELL-KW) dataset of 25 Subjects. PCA (Principal Component Analysis) and IQR (Interquartile Range) Preprocessing techniques are applied to select 26 features and detect outliers. The proposed model used a long short-term memory (LSTM) model to sort stress levels from biosensors in real-time and gives 98% accuracy. This study goes even further by using explainable artificial intelligence (XAI) models to explain their performance by pointing out the factors the model thought were important when making a decision. The SHAP (SHapley Additive Explanations) model is used to understand results by making them easier to interpret. It also promotes acknowledging stress as a beneficial method for managing mental health, highlighting the significance of early identification and intervention for a proactive and comprehensive approach to mental well-being. The contributions provide significant insights and techniques for resolving stress-related difficulties and developing mental health awareness and resilience.
Downloads
References
World employment and social outlook - Trends 2022. Geneva: International Labour Organization; 2022 (https://www.ilo.org/global/research/global-reports/weso/trends2022/WCMS_834081/lang--en/index.htm, accessed 26 August 2022)
Kirkbride JB, Anglin DM, Colman I, et al. The social determinants of mental health and disorder: evidence, prevention and recommendations. World Psychiatry. 2024;23(1):58-90. doi:10.1002/wps.21160
Paudel U, Parajuli A, Shrestha R, Kumari S, Adhikari Yadav S, Marahatta K. Perceived stress, sources of stress and coping strategies among undergraduate medical students of Nepal: a cross-sectional study. F1000Research. 2024 Jan 3;11:167.
Vasanthakumari S. Soft skills and its application in work place. World Journal of Advanced Research and Reviews. 2019;3(2):066-72.
Slavich GM, Roos LG, Mengelkoch S, et al. Social Safety Theory: Conceptual foundation, underlying mechanisms, and future directions. Health Psychol Rev. 2023;17(1):5-59. doi:10.1080/17437199.2023.2171900.
Mohammadi A, Fakharzadeh M, Baraeinejad B (2022) An integrated human stress detection sensor using supervised algorithms. IEEE Sens J 22(8):8216–8223. https:// doi. org/ 10. 1109/ jsen. 2022. 31577.
V. Motogna, G. Lupu-Florian and E. Lupu, "Strategy For Affective Computing Based on HRV and EDA," 2021 International Conference on e-Health and Bioengineering (EHB), Iasi, Romania, 2021, pp. 1-4, doi: 10.1109/EHB52898.2021.9657654.
M. Malik et al., "Heart rate variability: Standards of Measurement, Physiological," Eur. Heart J., vol. 17, pp. 354–381, 1996, [Online]. Available: http://www.mendeley.com/research/guidelines-heart-ratevariability-2/
Brennan M, Palaniswami M, Kamen P. Do existing measures of Poincare plot geometry reflect nonlinear features of heart rate variability?. IEEE transactions on biomedical engineering. 2001 Nov;48(11):1342-7.
J. Park and S. Kim, "Wearable Photoplethysmography Sensor," TENCON 2018 - 2018 IEEE Reg. 10 Conf., no. October, pp. 1899–1902, 2018.
T. Chalmers et al., "Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables," Sensors, vol. 22, no. 1, p. 151, Dec. 2021, DOI: 10.3390/s22010151.
K. M. Dalmeida and G. L. Masala, "HRV features as viable physiologicalmarkers for stress detection using wearable devices," Sensors, vol. 21, no.8, art. no. 2873, pp. 1–18, Mar. 2021.
State-of-the-Art of Stress Prediction from Heart Rate Variability Using Artificial Intelligence. Cognitive Computation. 16. 455–481. https://doi.org/10.1007/s12559-023-10200-0 Dahal, Kamana, Brian Bogue-Jimenez, and Ana Doblas. (2023). Global Stress Detection Framework Combining a Reduced Set of HRV Features and Random Forest Model. Sensors. 23. 11: 5220. https://doi.org/10.3390/s23115220.
Koldijk S, Sappelli M, Verberne S, Neerincx MA, Kraaij W. The swell knowledge work dataset for stress and user modeling research. InProceedings of the 16th international conference on multimodal interaction 2014 Nov 12 (pp. 291-298).
Fauquet-Alekhine P, Boucherand A. Psychological Assessment of the State of Mental Stress at Work. InThe Palgrave Handbook of Occupational Stress 2023 Aug 24 (pp. 97-115). Cham: Springer International Publishing.
K. Nkurikiyeyezu, A. Yokokubo, and G. Lopez, "The effect of person-specific biometrics in improving generic stress predictive models," arXiv, vol. 0, no. 0, pp. 1–12, 2019, doi: 10.18494/sam.2020.2650.
Y. S. Can, N. Chalabianloo, D. Ekiz, and C. Ersoy, "Continuous stress detection using wearable sensors in real life: Algorithmic programming contest case study," Sensors (Switzerland), vol. 19, no. 8, 2019, doi: 10.3390/s19081849.
D. Lee , "Long short-term memory recurrent neural network-based acoustic model using connectionist temporal classification on a large-scale training corpus," China Commun., vol. 14, no. 9, pp. 23–31, 2017, doi: 10.1109/CC.2017.8068761.
I. Yoshua Bengio, Patrice Simard, and Paolo Frasconi, Student Member, "Learning Long-Term Dependencies with Gradient Descent is Difficult," IEEE Trans. NEURAL NETWORKS, vol. 5, no. 2, pp. 157–166, 1994.
G. J. Sepp Hochreiter Fakult¨at f¨ur Informatik, Technische Universit¨at M¨unchen, 80290 M¨unchen and S. urgen Schmidhuber IDSIA, Corso Elvezia 36, 6900 Lugano, “Long Short-Term Memory,” Neural Comput. 9, vol. 50, no. 6, pp. 1735–1780 (1997), 1997, doi: 10.17582/journal.pjz/2018.50.6.2199.2207.
Ramteke, R.B., Thool, V.R. (2021). ECG Based Stress Detection in Automobile Drivers Using Long Short-Term Memory (LSTM) Network. In: Singh, M., Tyagi, V., Gupta, P.K., Flusser, J., Ören, T., Sonawane, V.R. (eds) Advances in Computing and Data Sciences. ICACDS 2021. Communications in Computer and Information Science, vol 1441. Springer, Cham. https://doi.org/10.1007/978-3-030-88244-0_32.
F. Pedregosa et al., "Scikit-learn: Machine Learning in Python," J. Mach. Learn. Res., vol. 12, pp. 2825– 2830, 2011.
A. Chatterjee, M. W. Gerdes, and S. G. Martinez, "Identification ofrisk factors associated with obesity and overweight – A machine learning overview," Sensors, vol. 20, no. 9, art. no. 2734, pp. 1–30, May 2020.
A. Chatterjee, N. Pahari, A. Prinz, and M. Reigler, "Machine learning and ontology in eCoaching for personalized activity level monitoring and recommendation generation," Scientific Reports, vol. 12, no. 1, pp. 1–26, Nov. 2022.
S. Sriramprakash, V. D. Prasanna, and O. V. R. Murthy, "Stress detectionin working people," Procedia Computer Sci., vol. 115, pp. 359–366, Oct.2017.
P. Sarkar and A. Etemad, "Self-supervised learning for ECG-based emotion recognition," in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process.(ICASSP), 2020, pp. 3217–3221.
S. Koldijk, M. Sappelli, S. Verberne, M. A. Neerincx, andW. Kraaij, "The SWELL knowledge work dataset for stress and user modeling research," in Proc. ACM Int. Conf. Multimodal Interaction (ICMI), 2014, pp. 291–298.
M. Albaladejo-Gonzalez, J. A. Ruiperez-Valiente, and F. Gomez Marmol, "Evaluating diferent confgurations of machine learning models andtheir transfer learning capabilities for stress detection using heart rate," J. Ambient Intell. Humanized Comput., Aug. 2022. [Online]. Available:https://doi.org/10.1007/s12652-022-04365-z.
Zawad, M.R.S., Rony, C.S.A., Haque, M.Y., Banna, M.H.A., Mahmud, M., Kaiser, M.S. (2023). A Hybrid Approach for Stress Prediction from Heart Rate Variability. In: Mandal, J.K., De, D. (eds) Frontiers of ICT in Healthcare . Lecture Notes in Networks and Systems, vol 519. Springer, Singapore. https://doi.org/10.1007/978-981-19-5191-6_10
Banerjee, J.S., Mahmud, M. & Brown, D. Heart Rate Variability-Based Mental Stress Detection: An Explainable Machine Learning Approach. SN COMPUT. SCI. 4, 176 (2023). https://doi.org/10.1007/s42979-022-01605-z
Chalabianloo N, Can YS, Umair M, Sas C, Ersoy C. Application level performance evaluation of wearable devices for stress classification with explainable AI. Pervasive and Mobile Computing. 2022 Dec 1;87:101703.
S. M. Lundberg and S. I. Lee, "A unified approach to interpreting model predictions," Adv. Neural Inf. Process. Syst., vol. 2017-Decem, no. December, pp. 4766–4775, 2017.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
