Predição de pressão sanguínea através de sinais de fotopletismografia usando redes neurais MLP e LSTM

Gustavo dos Santos Cardoso; Mateus Gomes Lucas; Samuel dos Santos Cardoso

doi:10.29384/rbfm.2021.v15.19849001637

Authors

Gustavo dos Santos Cardoso UNICAMP
Mateus Gomes Lucas UFSC
Samuel dos Santos Cardoso UFSC

DOI:

https://doi.org/10.29384/rbfm.2021.v15.19849001637

Keywords:

photoplethysmography, machine learning, blood pressure, LSTM

Abstract

Blood pressure is one of the basic vital signs of human beings and its measurement must be done regularly during the day by people who have some type of cardiovascular disease. Traditional methods of measurement usually need to occlude an artery, causing discomfort to the patient. This prevents continuous pressure monitoring and even discourages regular monitoring during the day. In view of this problem, in this work, two machine learning algorithms using photoplethysmography (PPG) signals are presented and compared to estimate blood pressure without the need for artery occlusion. The first algorithm implemented was a reproduction of the method proposed by Kurylyak et al., Which served as a reference for the development of the LSTM (Long Short-Term Memory) network proposed in this work, which used the temporal characteristics of the PPG signal as parameters input. For the training of the algorithms, the MIMIC II database (Multiparameter Intelligent Monitoring in Intensive Care) was used. For the evaluation of these methods, statistical analyzes of the results obtained were carried out in relation to the blood pressure reference values, already defined in the databases. The results obtained indicate that the architecture based on the LSTM network using temporal characteristics of the PPG signal as an input parameter can produce better results when compared to the MLP (Multi Layer Perceptron) network.

Downloads

Download data is not yet available.

References

1. Malachias MV, Plavnik FL, Machado CA, Malta D, Scala LC, Fuchs S. 7ª Diretriz Brasileira de Hipertensão Arterial: Capítulo 1-Conceituação, Epidemiologia e Prevenção Primária. Arquivos Brasileiros de Cardiologia. 2016 Sep;107(3):1-103.
2. Brasil. Ministério da Saúde. Cadernos de Atenção Básica – nº 15: Hipertensão Arterial Sistêmica. Brasil: 2006.
3. Hall JE. Guyton e Hall: Tratado de Fisiologia Médica. 13th ed Rio de Janeiro: Elsevier Editora Ltda. 2017:227-41.
4. Shioya K, Dohi T. Blood pressure measurement device based on the arterial tonometry method with micro triaxial force sensor. In2013 Transducers & Eurosensors XXVII: The 17th International Conference on Solid-State Sensors, Actuators and Microsystems (TRANSDUCERS & EUROSENSORS XXVII) 2013 Jun 16 (pp. 2389-2392). IEEE.
5. Kachuee M, Kiani MM, Mohammadzade H, Shabany M. Cuff-less high-accuracy calibration-free blood pressure estimation using pulse transit time. In2015 IEEE international symposium on circuits and systems (ISCAS) 2015 May 24 (pp. 1006-1009). IEEE.
6. Nabeel PM, Jayaraj J, Mohanasankar S. Single-source PPG-based local pulse wave velocity measurement: a potential cuffless blood pressure estimation technique. Physiological measurement. 2017 Nov 28;38(12):2122.
7. Kurylyak Y, Lamonaca F, Grimaldi D. A Neural Network-based method for continuous blood pressure estimation from a PPG signal. In2013 IEEE International instrumentation and measurement technology conference (I2MTC) 2013 May 6 (pp. 280-283). IEEE.
8. Xing X, Sun M. Optical blood pressure estimation with photoplethysmography and FFT-based neural networks. Biomedical optics express. 2016 Aug 1;7(8):3007-20.
9. Khalid SG, Zhang J, Chen F, Zheng D. Blood pressure estimation using photoplethysmography only: comparison between different machine learning approaches. Journal of healthcare engineering. 2018 Oct 23;2018.
10. Slapničar G, Mlakar N, Luštrek M. Blood pressure estimation from photoplethysmogram using a spectro-temporal deep neural network. Sensors. 2019 Jan;19(15):3420.
11. Association for the Advancement of Medical Instrumentation. American national standards for electronic or automated sphygmomanometers. ANSI/AAMI SP 10-1987. 1987.
12. Liu M, Po LM, Fu H. Cuffless blood pressure estimation based on photoplethysmography signal and its second derivative. International Journal of Computer Theory and Engineering. 2017 Jun 1;9(3):202.
13. Bersano N, Sanson H. Non-invasive blood pressure estimation from photoplethysmography signals using artificial neural networks. In2018 20th International Conference on Advanced Communication Technology (ICACT) 2018 Feb 11 (pp. 29-35). IEEE.
14. Kavsaoğlu AR, Polat K, Bozkurt MR. An innovative peak detection algorithm for photoplethysmography signals: an adaptive segmentation method. Turkish Journal of Electrical Engineering & Computer Sciences. 2016 Mar 23;24(3):1782-96.