Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation
Abstract
:1. Introduction
2. Related Work
3. Materials and Methods
3.1. Datasets
3.1.1. V4V Dataset
3.1.2. Operator Dataset
3.2. Proposed Approach
3.2.1. CNN Models Selection
3.2.2. ROI Selection
3.2.3. Building Our Models
3.2.4. Testing on Operator Dataset
3.2.5. Comparing with Other Models
4. Results
5. Discussion
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Wu, C.Y.; Hu, H.Y.; Chou, Y.J.; Huang, N.; Chou, Y.C.; Li, C.P. High Blood Pressure and All-Cause and Cardiovascular DiseaseMortalities in Community-Dwelling Older Adults. Medicine 2015, 94, e2160. [Google Scholar] [CrossRef] [PubMed]
- Jain, M.; Deb, S.; Subramanyam, A.V. Face video based touchless blood pressure and heart rate estimation. In Proceedings of the 2016 IEEE 18th International Workshop on Multimedia Signal Processing (MMSP), Montreal, QC, Canada, 21–23 September 2016; pp. 1–5. [Google Scholar]
- Secerbegovic, A.; Bergsland, J.; Halvorsen, P.S.; Suljanovic, N.; Mujcic, A.; Balasingham, I. Blood pressure estimation using video plethysmography. In Proceedings of the 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI), Prague, Czech Republic, 13–16 April 2016; pp. 461–464. [Google Scholar]
- Castaneda, D.; Esparza, A.; Ghamari, M.; Soltanpur, C.; Nazeran, H. A review on wearable photoplethysmography sensors and their potential future applications in health care. Int. J. Biosens. Bioelectron. 2018, 4, 195–202. [Google Scholar]
- O’Shea, K.; Nash, R. An Introduction to Convolutional Neural Networks. arXiv 2015, arXiv:1511.08458. Available online: https://arxiv.org/abs/1511.08458 (accessed on 16 December 2021).
- Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [Google Scholar] [CrossRef] [PubMed]
- Verkruysse, W.; Svaasand, L.O.; Nelson, J.S. Remote plethysmographic imaging using ambient light. Opt. Express 2008, 16, 21434–21445. [Google Scholar] [CrossRef] [PubMed]
- Luo, H.; Yang, D.; Barszczyk, A.; Vempala, N.; Wei, J.; Wu, S.J.; Zheng, P.P.; Fu, G.; Lee, K.; Feng, Z.P. Smartphone-Based Blood Pressure Measurement Using Transdermal Optical Imaging Technology. Circ. Cardiovasc. Imaging 2019, 12, e008857. [Google Scholar] [CrossRef] [PubMed]
- Hyvärinen, A.; Hurri, J.; Hoyer, P.O. Independent Component Analysis. In Natural Image Statistics: A Probabilistic Approach to Early Computational Vision; Springer: London, UK, 2009; pp. 151–175. [Google Scholar]
- Oiwa, K.; Bando, S.; Nozawa, A. Contactless blood pressure sensing using facial visible and thermal images. Artif. Life Robot. 2018, 23, 387–394. [Google Scholar] [CrossRef]
- Iuchi, K.; Miyazaki, R.; Cardoso, G.C.; Ogawa-Ochiai, K.; Tsumura, N. Remote Estimation of Continuous Blood Pressure by a Convolutional Neural Network Trained on Spatial Patterns of Facial Pulse Waves. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, New Orleans, LA, USA, 19–24 June 2022; pp. 2139–2145. [Google Scholar]
- Wu, B.F.; Chiu, L.W.; Wu, Y.C.; Lai, C.C.; Chu, P.H. Contactless Blood Pressure Measurement via Remote Photoplethysmography With Synthetic Data Generation Using Generative Adversarial Network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, New Orleans, LA, USA, 18–24 June 2022; pp. 2130–2138. [Google Scholar]
- De Haan, G.; Jeanne, V. Robust pulse rate from chrominance-based rPPG. IEEE Trans. Biomed. Eng. 2013, 60, 2878–2886. [Google Scholar] [CrossRef] [PubMed]
- Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative Adversarial Networks. Commun. ACM 2020, 63, 139–144. [Google Scholar] [CrossRef]
- Schrumpf, F.; Frenzel, P.; Aust, C.; Osterhoff, G.; Fuchs, M. Assessment of Non-Invasive Blood Pressure Prediction from PPG and rPPG Signals Using Deep Learning. Sensors 2021, 21, 6022. [Google Scholar] [CrossRef] [PubMed]
- Krizhevsky, A.; Sutskever, I.; Hinton, G.E. ImageNet classification with deep convolutional neural networks. Adv. Neural Inf.Process. Syst. 2017, 60, 84–90. [Google Scholar] [CrossRef]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep Residual Learning for Image Recognition. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778. [Google Scholar]
- Slapničar, G.; Mlakar, N.; Luštrek, M. Blood pressure estimation from photoplethysmogram using a spectro-temporal deep neural network. Sensors 2019, 19, 3420. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Wang, W.; Den Brinker, A.C.; Stuijk, S.; De Haan, G. Algorithmic principles of remote PPG. IEEE Trans. Biomed. Eng. 2017, 64, 1479–1491. [Google Scholar] [CrossRef] [PubMed]
- Jeong, I.C.; Finkelstein, J. Introducing contactless blood pressure assessment using a high speed video camera. J. Med. Syst. 2016, 40, 77. [Google Scholar] [CrossRef] [PubMed]
- Finkelstein, J.; Jeong, I.C. Towards Contactless Monitoring of Blood Pressure at Rest and During Exercise Using Infrared Imaging. In Proceedings of the 2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York, NY, USA, 28–31 October 2020; pp. 756–758. [Google Scholar]
- Gaurav, A.; Maheedhar, M.; Tiwari, V.N.; Narayanan, R. Cuff-less PPG based continuous blood pressure monitoring—A smartphone based approach. In Proceedings of the 2016 38th annual international conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 16–20 August 2016; pp. 607–610. [Google Scholar]
- Visvanathan, A.; Sinha, A.; Pal, A. Estimation of blood pressure levels from reflective photoplethysmograph using smart phones. In Proceedings of the 13th IEEE International Conference on BioInformatics and BioEngineering, Chania, Greece, 10–13 November 2013; pp. 1–5. [Google Scholar]
- Revanur, A.; Li, Z.; Ciftci, U.A.; Yin, L.; Jeni, L.A. The First Vision for Vitals (V4V) Challenge for Non-Contact Video-Based Physiological Estimation. In Proceeding of the 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), Montreal, BC, Canada, 11–17 October 2021; pp. 2760–2767. [Google Scholar]
- Zhang, Z.; Girard, J.M.; Wu, Y.; Zhang, X.; Liu, P.; Ciftci, U.; Canavan, S.; Reale, M.; Horowitz, A.; Yang, H.; et al. Multimodal spontaneous emotion corpus for human behavior analysis. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 3438–3446. [Google Scholar]
- InformedHealth.org. What Is Blood Pressure and How Is It Measured? Available online: https://www.ncbi.nlm.nih.gov/books/NBK279251/ (accessed on 29 November 2022).
- Smulyan, H.; Safar, M.E. The diastolic blood pressure in systolic hypertension. Ann. Intern. Med. 2000, 132, 233–237. [Google Scholar] [CrossRef] [PubMed]
- Herakova, N.; Nwobodo, N.H.N.; Wang, Y.; Chen, F.; Zheng, D. Effect of respiratory pattern on automated clinical blood pressure measurement: An observational study with normotensive subjects. Clin. Hypertens. 2017, 23, 15. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Saha, R. Transfer Learning—A Comparative Analysis. Bachelor’s Thesis, Kalinga Institute of Industrial Technology, Bhubaneswar, India, December 2018. Available online: https://www.researchgate.net/publication/329786975_Transfer_Learning_-_A_Comparative_Analysis (accessed on 1 December 2022).
- Keras Applications. Available online: https://keras.io/api/applications/ (accessed on 12 August 2022).
- Guo, J.; Zhu, X.; Yang, Y.; Yang, F.; Lei, Z.; Li, S.Z. Towards fast, accurate and stable 3d dense face alignment. In Proceedings of the European Conference on Computer Vision, Glasgow, UK, 23–28 August 2020; pp. 152–168. [Google Scholar]
- Pearson Product-Moment Correlation. Available online: https://statistics.laerd.com/statistical-guides/pearson-correlation-coefficient-statistical-guide.php (accessed on 3 December 2022).
- Schrumpf, F. Fabian-Sc85/Non-Invasive-bp-Estimation-Using-Deep-Learning. Available online: https://github.com/Fabian-Sc85/Non-Invasive-bp-Estimation-Using-Deep-Learning (accessed on 24 October 2022).
- Dietterich, T.G. Ensemble methods in machine learning. In Proceedings of the International Workshop on Multiple Classifier Systems, Cagliari, Italy, 21–23 June 2000; pp. 1–15. [Google Scholar]
- Nguyen, J.D.; Duong, H. Anatomy, Head and Neck, Cheeks. Available online: https://www.ncbi.nlm.nih.gov/books/NBK546659/ (accessed on 7 December 2022).
Pretrained Model | RMSE | MAE |
---|---|---|
Xception | 24.325 | 21.679 |
VGG16 | 21.142 | 17.178 |
Resnet50V2 | 17.427 | 14.883 |
InceptionV3 | 22.648 | 18.953 |
DenseNet121 | 22.358 | 18.267 |
EfficientNetB0 | 16.652 | 14.237 |
EfficientNetB1 | 16.333 | 14.006 |
EfficientNetB2 | 15.958 | 13.386 |
ROI | RMSE | MAE |
---|---|---|
Forehead | 24.252 | 21.218 |
Left cheek | 17.884 | 15.063 |
Right cheek | 16.332 | 14.773 |
Model Type | Number of the Model | MAE | Mean Accuracy |
---|---|---|---|
SBP | Model_1 | 12.408 | 89.1% |
Model_2 | 14.491 | 87.3% | |
Model_3 | 14.174 | 87.8% | |
DBP | Model_1 | 12.652 | 83.4% |
Model_2 | 11.371 | 85.2% | |
Model_3 | 16.071 | 79.5% |
Combination Type | Included Models | MAE | Mean Accuracy (%) |
---|---|---|---|
SBP | Model_1, Model_2 | 11.867 | 89.5% |
Model_1, Model_3 | 11.976 | 89.6% | |
DBP | Model_1, Model 2 | 10.706 | 86.2% |
Model_1, Model_3 | 12.499 | 84.1% |
Model Type | Number of the Model | MAE | Mean Accuracy (%) |
---|---|---|---|
SBP | AlexNet [16] | 18.582 | 81.8% |
ResNet [17] | 15.456 | 85.4% | |
LSTM | 14.579 | 86.7% | |
Slapnicar et al. [18] | 31.359 | 64.3% | |
Combination (Model_1, Model 2) | 17.477 | 84.7% | |
Combination (Model_1, Model 3) | 15.121 | 86.7% | |
Model_1 | 18.084 | 83.7% | |
Model_2 | 16.424 | 84.4% | |
Model_3 | 13.749 | 88.2% | |
DBP | AlexNet [16] | 13.460 | 83.1% |
ResNet [17] | 12.528 | 83.9% | |
LSTM | 13.119 | 83.5% | |
Slapnicar et al. [18] | 20.621 | 62.1% | |
Combination (Model_1, Model 2) | 12.012 | 84.04% | |
Combination (Model_1, Model 3) | 11.169 | 84.9% | |
Model_1 | 12.391 | 83.8% | |
Model_2 | 12.697 | 83.02% | |
Model_3 | 14.833 | 79.2% |
Model Type | Number of the Model | Average Pearson’s Correlation Coefficient over 60 Videos | Average p-Value over 60 Videos |
---|---|---|---|
SBP | AlexNet [16] | 0.4256 | 0.124 |
ResNet [17] | 0.4226 | 0.129 | |
LSTM | 0.4189 | 0.133 | |
Slapnicar et al. [18] | - | - | |
Combination (Model_1, Model 2) | 0.4606 | 0.107 | |
Combination (Model_1, Model 3) | 0.4863 | 0.092 | |
Model_1 | 0.4670 | 0.103 | |
Model_2 | 0.5028 | 0.048 | |
Model_3 | 0.4304 | 0.122 | |
DBP | AlexNet [16] | 0.4491 | 0.118 |
ResNet [17] | 0.4559 | 0.112 | |
LSTM | 0.3920 | 0.154 | |
Slapnicar et al. [18] | - | - | |
Combination (Model_1, Model 2) | 0.5236 | 0.045 | |
Combination (Model_1, Model 3) | 0.4623 | 0.106 | |
Model_1 | 0.5302 | 0.037 | |
Model_2 | 0.4743 | 0.097 | |
Model_3 | 0.4429 | 0.120 |
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. |
© 2023 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
Hamoud, B.; Kashevnik, A.; Othman, W.; Shilov, N. Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation. Sensors 2023, 23, 1753. https://doi.org/10.3390/s23041753
Hamoud B, Kashevnik A, Othman W, Shilov N. Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation. Sensors. 2023; 23(4):1753. https://doi.org/10.3390/s23041753
Chicago/Turabian StyleHamoud, Batol, Alexey Kashevnik, Walaa Othman, and Nikolay Shilov. 2023. "Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation" Sensors 23, no. 4: 1753. https://doi.org/10.3390/s23041753
APA StyleHamoud, B., Kashevnik, A., Othman, W., & Shilov, N. (2023). Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation. Sensors, 23(4), 1753. https://doi.org/10.3390/s23041753