Assessment of Mental Workload Level Based on PPG Signal Fusion Continuous Wavelet Transform and Cardiopulmonary Coupling Technology
Abstract
:1. Introduction
- In this paper, we propose a new interpretable pre-processing feature transformation method for mental load classification.
- This paper compares multiple feature extraction methods commonly used in the past on the same dataset to obtain a relatively objective conclusion.
- The method proposed in this paper achieves a performance that approximates the performance of using ECG signals when only PPG signals are used.
2. Materials and Methods
2.1. Database
2.2. Signal Pre-Processing
- (1)
- Signal filtering
- (2)
- Derived sequence extraction
- (3)
- PPG feature extraction
- (4)
- Signal Transformation
2.3. Machine and Deep Learning Model
2.4. Performance Evaluation
3. Results
4. Discussion
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Appendix A. HRV Metrics Covered in This Article and Detailed Description
Feature Type | Features | Detailed Description of the Features |
Time domain | mean_nni | Mean of the normal-to-normal intervals (NNIs) |
sdnn | Standard deviation of the NNIs, indicating overall HRV | |
sdsd | Standard deviation of successive NNI differences | |
nni_50 | Number of pairs of successive NNIs that differ by more than 50 ms | |
pnni_50 | Percentage of NNI pairs differing by more than 50 ms | |
nni_20 | Number of pairs of successive NNIs that differ by more than 20 ms | |
pnni_20 | Percentage of NNI pairs differing by more than 20 ms | |
rmssd | Root mean square of successive NNI differences | |
median_nni | Median of NNIs indicating the central tendency of HRV values | |
std_hr | Standard deviation of heart rate values | |
range_nni | Difference between the maximum and minimum NNIs | |
cssd | Corrected sum of squares of successive NNI differences | |
cvnni | Coefficient of variation for NNIs | |
mean_hr | Mean of heart rate values representing average heart rate | |
max_hr | Maximum heart rate value observed during the recording period | |
min_hr | Minimum heart rate value observed during the recording period | |
Frequency domain | lf | Low-frequency power, representing the contribution of sympathetic and parasympathetic activities |
hf | High-frequency power, indicating the parasympathetic activity | |
lf_hf_ratio | Ratio of low-frequency power to high-frequency power | |
lfnu | Normalized low-frequency power | |
hfnu | Normalized high-frequency power | |
total_power | Total power in the HRV signal | |
vlf | Very low-frequency power reflecting long-term regulatory mechanisms | |
Nonlinear domain | sd1 | Standard deviation of the points perpendicular to the line of identity in the Poincaré plot |
sd2 | Standard deviation of the points along the line of identity in the Poincaré plot | |
ratio_sd2_sd1 | Ratio of sd2 to sd1 | |
sampen | Sample entropy, measuring the complexity or irregularity of the HRV signal | |
csi | Cardiac Sympathetic Index | |
cvi | Cardiac Vagal Index | |
Modified_csi | Modified Cardiac Sympathetic Index | |
Geometric domain | triangular_index | Quantifying the distribution of RR intervals |
References
- Dehais, F.; Lafont, A.; Roy, R.; Fairclough, S. A Neuroergonomics Approach to Mental Workload, Engagement and Human Performance. Front. Neurosci. 2020, 14, 268. [Google Scholar] [CrossRef]
- Qiu, H.; Li, M.; Bai, B.; Wang, N.; Li, Y. The impact of AI-enabled service attributes on service hospitableness: The role of employee physical and psychological workload. Int. J. Contemp. Hosp. Manag. 2022, 34, 1374–1398. [Google Scholar] [CrossRef]
- Cheng, B.; Fan, C.; Fu, H.; Huang, J.; Chen, H.; Luo, X. Measuring and computing cognitive statuses of construction workers based on electroencephalogram: A critical review. IEEE Trans. Comput. Soc. Syst. 2022, 9, 1644–1659. [Google Scholar] [CrossRef]
- Gevins, A.; Smith, M.E.; Leong, H.; McEvoy, L.; Whitfield, S.; Du, R.; Rush, G. Monitoring Working Memory Load during Computer-Based Tasks with EEG Pattern Recognition Methods. Hum. Factors J. Hum. Factors Ergon. Soc. 1998, 40, 79–91. [Google Scholar] [CrossRef]
- Iwanaga, K.; Saito, S.; Shimomura, Y.; Harada, H.; Katsuura, T. The Effect of Mental Loads on Muscle Tension, Blood Pressure and Blink Rate. J. Physiol. Anthropol. Appl. Hum. Sci. 2000, 19, 135–141. [Google Scholar] [CrossRef]
- Salahuddin, L.; Cho, J.; Jeong, M.G.; Kim, D. Ultra Short Term Analysis of Heart Rate Variability for Monitoring Mental Stress in Mobile Settings. In Proceedings of the 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Lyon, France, 22–26 August 2007; pp. 4656–4659. [Google Scholar]
- Tamantini, C.; Rondoni, C.; Cordella, F.; Guglielmelli, E.; Zollo, L. A Classification Method for Workers’ Physical Risk. Sensors 2023, 23, 1575. [Google Scholar] [CrossRef]
- Xuan, Q.; Wu, J.; Shen, J.; Ji, X.; Lyu, Y.; Zhang, Y. Assessing cognitive load in adolescent and adult students using photoplethysmogram morphometrics. Cogn. Neurodynamics 2020, 14, 709–721. [Google Scholar] [CrossRef]
- Vicente-Samper, J.M.; Tamantini, C.; Ávila-Navarro, E.; De La Casa-Lillo, M.; Zollo, L.; Sabater-Navarro, J.M.; Cordella, F. An ML-Based Approach to Reconstruct Heart Rate from PPG in Presence of Motion Artifacts. Biosensors 2023, 13, 718. [Google Scholar] [CrossRef]
- Wang, Z.-H.; Wu, Y.-C. A Novel Rapid Assessment of Mental Stress by Using PPG Signals Based on Deep Learning. IEEE Sens. J. 2022, 22, 21232–21239. [Google Scholar] [CrossRef]
- Cinaz, B.; Arnrich, B.; La Marca, R.; Tröster, G. Monitoring of mental workload levels during an everyday life office-work scenario. Pers. Ubiquitous Comput. 2013, 17, 229–239. [Google Scholar] [CrossRef]
- Zhao, L.; Niu, X.; Wang, L.; Niu, J.; Zhu, X.; Dai, Z. Stress Detection via Multimodal Multitemporal-Scale Fusion: A Hybrid of Deep Learning and Handcrafted Feature Approach. IEEE Sens. J. 2023, 23, 27817–27827. [Google Scholar] [CrossRef]
- Zhang, X.; Lyu, Y.; Hu, X.; Hu, Z.; Shi, Y.; Yin, H. Evaluating Photoplethysmogram as a Real-Time Cognitive Load Assessment during Game Playing. Int. J. Hum.–Comput. Interact. 2018, 34, 695–706. [Google Scholar] [CrossRef]
- Amin, M.; Ullah, K.; Asif, M.; Waheed, A.; Haq, S.U.; Zareei, M.; Biswal, R.R. ECG-Based Driver’s Stress Detection Using Deep Transfer Learning and Fuzzy Logic Approaches. IEEE Access 2022, 10, 29788–29809. [Google Scholar] [CrossRef]
- Zubair, M.; Yoon, C. Multilevel mental stress detection using ultra-short pulse rate variability series. Biomed. Signal Process. Control 2020, 57, 101736. [Google Scholar] [CrossRef]
- Beh, W.-K.; Wu, Y.-H.; Wu, A.-Y. Robust PPG-Based Mental Workload Assessment System Using Wearable Devices. IEEE J. Biomed. Health Inform. 2023, 27, 2323–2333. [Google Scholar] [CrossRef] [PubMed]
- Thomas, R.J.; Mietus, J.E.; Peng, C.-K.; Goldberger, A.L. An Electrocardiogram-Based Technique to Assess Cardiopulmonary Coupling During Sleep. Sleep 2005, 28, 1151–1161. [Google Scholar] [CrossRef] [PubMed]
- Beh, W.-K.; Wu, Y.-H.; Wu, A.-Y. MAUS: A Dataset for Mental Workload Assessment on N-Back Task Using Wearable Sensor. arXiv 2021, arXiv:2111.025612021. [Google Scholar]
- Schäfer, A.; Vagedes, J. How accurate is pulse rate variability as an estimate of heart rate variability? Int. J. Cardiol. 2013, 166, 15–29. [Google Scholar] [CrossRef]
- Castaldo, R.; Montesinos, L.; Melillo, P.; James, C.; Pecchia, L. Ultra-short term HRV features as surrogates of short term HRV: A case study on mental stress detection in real life. BMC Med. Inform. Decis. Mak. 2019, 19, 12. [Google Scholar] [CrossRef]
- Scholkmann, F.; Boss, J.; Wolf, M. An Efficient Algorithm for Automatic Peak Detection in Noisy Periodic and Quasi-Periodic Signals. Algorithms 2012, 5, 588–603. [Google Scholar] [CrossRef]
- Charlot, K.; Cornolo, J.; Brugniaux, J.V.; Richalet, J.P.; Pichon, A. Interchangeability between heart rate and photoplethysmography variabilities during sympathetic stimulations. Physiol. Meas. 2009, 30, 1357–1369. [Google Scholar] [CrossRef] [PubMed]
- Taelman, J.; Vandeput, S.; Spaepen, A.; Van Huffel, S. Influence of Mental Stress on Heart Rate and Heart Rate Variability. In Proceedings of the 4th European Conference of the International Federation for Medical and Biological Engineering, Antwerp, Belgium, 23–27 November 2008; Vander Sloten, J., Verdonck, P., Nyssen, M., Haueisen, J., Eds.; Springer: Berlin/Heidelberg, Germany, 2009; pp. 1366–1369. [Google Scholar]
- Díaz-García, J.; González-Ponce, I.; Ponce-Bordón, J.C.; López-Gajardo, M.Á.; Ramírez-Bravo, I.; Rubio-Morales, A.; García-Calvo, T. Mental Load and Fatigue Assessment Instruments: A Systematic Review. Int. J. Environ. Res. Public Health 2021, 19, 419. [Google Scholar] [CrossRef] [PubMed]
- Daniela, M.; Catalina, L.; Ilie, O.; Paula, M.; Daniel-Andrei, I.; Ioana, B. Effects of Exercise Training on the Autonomic Nervous System with a Focus on Anti-Inflammatory and Antioxidants Effects. Antioxidants 2022, 11, 350. [Google Scholar] [CrossRef]
- Fiľo, P.; Janoušek, O. The relation between physical and mental load, and the course of physiological functions and cognitive performance. Theor. Issues Ergon. Sci. 2022, 23, 38–59. [Google Scholar] [CrossRef]
- Jaiswal, D.; Chowdhury, A.; Banerjee, T.; Chatterjee, D. Effect of Mental Workload on Breathing Pattern and Heart Rate for a Working Memory Task: A Pilot Study. In Proceedings of the 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, 23–27 July 2019; pp. 2202–2206. [Google Scholar]
- Guo, D.; Peng, C.-K.; Wu, H.-L.; Mietus, J.E.; Liu, Y.; Sun, R.-S.; Thomas, R.J. ECG-derived cardiopulmonary analysis of pediatric sleep-disordered breathing. Sleep Med. 2011, 12, 384–389. [Google Scholar] [CrossRef] [PubMed]
- Cheng, P.; Chen, Z.; Li, Q.; Gong, Q.; Zhu, J.; Liang, Y. Atrial Fibrillation Identification with PPG Signals Using a Combination of Time-Frequency Analysis and Deep Learning. IEEE Access 2020, 8, 172692–172706. [Google Scholar] [CrossRef]
- Handelman, G.S.; Kok, H.K.; Chandra, R.V.; Razavi, A.H.; Lee, M.J.; Asadi, H. eDoctor: Machine learning and the future of medicine. J. Intern. Med. 2018, 284, 603–619. [Google Scholar] [CrossRef] [PubMed]
- Hu, X.; Yin, S.; Zhang, X.; Menon, C.; Fang, C.; Chen, Z.; Elgendi, M.; Liang, Y. Blood pressure stratification using photoplethysmography and light gradient boosting machine. Front. Physiol. 2023, 14, 1072273. [Google Scholar] [CrossRef]
- Esgalhado, F.; Batista, A.; Vassilenko, V.; Russo, S.; Ortigueira, M. Peak Detection and HRV Feature Evaluation on ECG and PPG Signals. Symmetry 2022, 14, 1139. [Google Scholar] [CrossRef]
- Schaule, F.; Johanssen, J.O.; Bruegge, B.; Loftness, V. Employing Consumer Wearables to Detect Office Workers’ Cognitive Load for Interruption Management. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 2018, 2, 32:1–32:20. [Google Scholar] [CrossRef]
- Gurel, N.Z.; Jung, H.; Hersek, S.; Inan, O.T. Fusing Near-Infrared Spectroscopy with Wearable Hemodynamic Measurements Improves Classification of Mental Stress. IEEE Sens. J. 2019, 19, 8522–8531. [Google Scholar] [CrossRef]
- Rashid, N.; Chen, L.; Dautta, M.; Jimenez, A.; Tseng, P.; Al Faruque, M.A. Feature Augmented Hybrid CNN for Stress Recognition Using Wrist-based Photoplethysmography Sensor. arXiv 2021, arXiv:2108.03166. [Google Scholar]
- Morelli, D.; Rossi, A.; Cairo, M.; Clifton, D.A. Analysis of the Impact of Interpolation Methods of Missing RR-intervals Caused by Motion Artifacts on HRV Features Estimations. Sensors 2019, 19, 3163. [Google Scholar] [CrossRef]
Feature Type | Features |
---|---|
TimeDomain (16) | mean_nni, sdnn, sdsd, nni_50, pnni_50, nni_20, pnni_20, rmssd, median_nni, std_hr, range_nni, cssd, cvnni, mean_hr, max_hr, min_hr |
FrequencyDomain (7) | lf, hf, lf_hf_ratio, lfnu, hfnu, total_power, vlf |
NonlinearDomain (7) | sd1, sd2, ratio_sd2_sd1, sampen, csi, cvi, Modified_csi |
GeometricDomain | triangular_index |
Layers | Output Size | Number of Filters | Feature Maps |
---|---|---|---|
Input | 120 × 1 | - | - |
1D convolution (C1) | (118, 5) | 5 | 118 × 5 |
1D max pooling (P1) | (116, 5) | - | 116 × 5 |
Dropout (DP1) | (116, 5) | - | 116 × 5 |
1D convolution (C2) | (114, 5) | 5 | 114 × 5 |
1D max pooling (P2) | (112, 5) | - | 112 × 5 |
Dropout (DP2) | (112, 5) | - | 112 × 5 |
1D convolution (C3) | (110, 5) | 5 | 110 × 5 |
1D max pooling (P3) | (108, 5) | - | 108 × 5 |
Dropout (DP3) | (108, 5) | - | 108 × 5 |
LSTM1 | (108, 64) | - | 108 × 64 |
LSTM2 | (64,) | - | 64 × 1 |
Fully connected (FC) | (2,) | - | 2 × 1 |
Classifier (sigmoid activation) | (2,) | - | 2 × 1 |
Kernel | Regularization C | Class | Class Weight |
---|---|---|---|
Linear | (10 × 10−3~300) | 2 | (0:0.67, 1:0.33) |
Objective | Learning Rate | Max Depth | Alpha | Class Weight | |
---|---|---|---|---|---|
XGBoost | Binary/logistic | (1 × 10−8, 1.0, ‘log-uniform’) | (2, 20, ‘int’) | (0, 10, ‘int’) | (1, 2, ‘int’) |
LightGBM | Binary/gbdt | (1 × 10−8, 1.0, ‘log-uniform’) | (2, 20, ‘int’) | (0, 10, ‘int’) | (1, 2, ‘int’) |
Signal Transformation | Models | Input | ACC (%) | F1 (%) | SP (%) | SN (%) | PRE (%) | AUC (%) | |
---|---|---|---|---|---|---|---|---|---|
Trial I | Derived sequence | 1DCNN +LSTM | PPG | 69.37 | 71.98 | 69.44 | 69.32 | 74.85 | 69.38 |
PPG (IBI) | 73.27 | 75.06 | 78.82 | 69.27 | 81.91 | 74.04 | |||
Trial II | Handmade features | SVM | PRV | 61.54 | 62.18 | 59.32 | 63.79 | 60.66 | 61.56 |
XGBoost | PRV | 68.75 | 72.22 | 60.81 | 75.58 | 69.15 | 68.2 | ||
LightGBM | PRV | 69.23 | 64 | 83.05 | 55.17 | 76.19 | 69.11 | ||
Trial III | Image | ResAtt Net | PPG (CWT) | 77.44 | 84.08 | 69.85 | 80.05 | 88.55 | 74.95 |
PPG (CPC+CWT) | 80.47 | 86.52 | 79.54 | 80.74 | 93.19 | 80.14 |
Signal Transformation | Models | Input | ACC (%) | F1 (%) | SP (%) | SN (%) | PRE (%) | AUC (%) | |
---|---|---|---|---|---|---|---|---|---|
Trial I | Derived sequence | 1D-ResNet +LSTM | ECG | 76.73 | 78.27 | 78.59 | 75.25 | 81.55 | 76.92 |
ECG (IBI) | 78.21 | 81.45 | 84.99 | 74.44 | 89.91 | 79.72 | |||
Trial II | Handmade features | SVM | HRV | 73.91 | 72.73 | 90 | 61.54 | 88.89 | 75.77 |
PRV+HRV | 75.63 | 78.69 | 66.22 | 83.72 | 74.23 | 75.77 | |||
XGBoost | HRV | 69.53 | 72.73 | 58.73 | 80 | 66.67 | 69.37 | ||
PRV+HRV | 79.91 | 79.81 | 79.63 | 80.19 | 79.44 | 79.91 | |||
LightGBM | HRV | 70.31 | 71.21 | 68.25 | 72.31 | 70.15 | 70.28 | ||
PRV+HRV | 78.04 | 73.14 | 93.33 | 61.75 | 89.68 | 77.54 | |||
Trial III | Image | ResAtt Net | ECG (CWT) | 77.51 | 84.65 | 74.62 | 78.26 | 92.18 | 76.44 |
ECG (CWT+CPC) | 82.41 | 87.39 | 77.01 | 84.49 | 90.49 | 80.75 |
Articles | Features | Task | Methods | ACC |
---|---|---|---|---|
Cinaz et al. (2011) [11] | ECG (HRV) | Officeworks | LDA, KNN, SVM | 71% |
Schedule et al. (2018) [33] | PPG (HRV) | N-back | SVM, Random Forest | 66% |
Gurel et al. (2019) [34] | PPG+ECG+SCG | N-back | ANOVA+RF | 85% |
Beh et al. (2021) [16] | PPG (HRV) | N-back (MAUS) | Linear SVM | 74% |
Rashid et al. (2021) [35] | PPG | Driver stress test (WESAD) | Hybrid CNN | 75% |
This study | PPG (CWT) | N-back (MAUS) | ResAttNet | 77% |
PPG (CWT+CPC) | N-back (MAUS) | ResAttNet | 80% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, H.; Wang, Z.; Zhuang, Y.; Yin, S.; Chen, Z.; Liang, Y. Assessment of Mental Workload Level Based on PPG Signal Fusion Continuous Wavelet Transform and Cardiopulmonary Coupling Technology. Electronics 2024, 13, 1238. https://doi.org/10.3390/electronics13071238
Zhang H, Wang Z, Zhuang Y, Yin S, Chen Z, Liang Y. Assessment of Mental Workload Level Based on PPG Signal Fusion Continuous Wavelet Transform and Cardiopulmonary Coupling Technology. Electronics. 2024; 13(7):1238. https://doi.org/10.3390/electronics13071238
Chicago/Turabian StyleZhang, Han, Ziyi Wang, Yan Zhuang, Shimin Yin, Zhencheng Chen, and Yongbo Liang. 2024. "Assessment of Mental Workload Level Based on PPG Signal Fusion Continuous Wavelet Transform and Cardiopulmonary Coupling Technology" Electronics 13, no. 7: 1238. https://doi.org/10.3390/electronics13071238
APA StyleZhang, H., Wang, Z., Zhuang, Y., Yin, S., Chen, Z., & Liang, Y. (2024). Assessment of Mental Workload Level Based on PPG Signal Fusion Continuous Wavelet Transform and Cardiopulmonary Coupling Technology. Electronics, 13(7), 1238. https://doi.org/10.3390/electronics13071238