ReliefF Feature Selection and a SSA-IRF Model for Continuous Blood Pressure Estimation from Electrocardiogram and Photoplethysmography Signals

Yunfei Ma; Yang Xiao; Wang Zhang

doi:10.54097/dgcw7275

Authors

Yunfei Ma
Yang Xiao
Wang Zhang

DOI:

https://doi.org/10.54097/dgcw7275

Keywords:

Continuous Blood Pressure Monitoring, Photoplethysmography (PPG), Electrocardiogram (ECG), Feature Selection, ReliefF Algorithm, Sparrow Search Algorithm (SSA), Iterative Random Forest (IRF)

Abstract

Blood pressure (BP) serves as a vital indicator of cardiovascular health; however, achieving accurate and continuous BP monitoring continues to present significant challenges. To enhance the performance of continuous BP estimation using electrocardiogram (ECG) and photoplethysmography (PPG) signals, this paper introduces a method that integrates the ReliefF feature weighting algorithm with a Sparrow Search Algorithm-optimized Iterative Random Forest (SSA-IRF) regression model. First, a comprehensive set of time-domain, frequency-domain, and time-frequency features is extracted from PPG and ECG signals. The ReliefF algorithm is then applied to select highly sensitive features strongly correlated with BP, thereby reducing redundancy and improving monitoring efficiency. Subsequently, a hybrid prediction model is developed by combining the Sparrow Search Algorithm (SSA) with an Iterative Random Forest (IRF) regression model to learn the mapping between the selected features and BP values. A novel fitness function is designed to balance prediction accuracy and consistency. Finally, ablation and comparative experiments were conducted using ECG and PPG signals from 200 subjects in the MIMIC-III database, validating the effectiveness and advancement of the proposed approach. Experimental results show that the method achieves a mean absolute error (MAE) of 2.73 mmHg and a standard deviation (STD) of 3.88 mmHg for systolic BP prediction, and an MAE of 1.62 mmHg with an STD of 2.31 mmHg for diastolic BP prediction. Its performance not only complies with the Association for the Advancement of Medical Instrumentation (AAMI) standards but also meets Grade A criteria according to the British Hypertension Society (BHS) protocol.

Downloads

Download data is not yet available.

References

[1] Mills, K. T., Stefanescu, A., & He, J. (2020). The global epidemiology of hypertension. NATURE REVIEWS NEPHROLOGY, 16(4), 223-237. https://doi.org/10. 1038/s 41581- 019-0244-2.

[2] Schutte, A. E., Kollias, A., & Stergiou, G. S. (2022). Blood pressure and its variability: classic and novel measurement techniques. NATURE REVIEWS CARDIOLOGY, 19(10), 643-654. https://doi.org/10.1038/s41569-022-00690-0.

[3] Mukkamala, R., Stergiou, G. S., & Avolio, A. P. (2022). Cuffless Blood Pressure Measurement. ANNUAL REVIEW OF BIOMEDICAL ENGINEERING, 24, 203-230. https:// doi. org/ 10.1146/annurev-bioeng-110220-014644.

[4] Poon, C. C. Y., Zhang, Y. T., & Ieee. (2005). Cuff-less and noninvasive measurements of arterial blood pressure by pulse transit time 2005 27TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-7.

[5] Kachuee, M., Kiani, M. M., Mohammadzade, H., & Shabany, M. (2017). Cuffless Blood Pressure Estimation Algorithms for Continuous Health-Care Monitoring. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 64(4), 859-869. https:// doi. org/10.1109/TBME.2016.2580904.

[6] Mancia, G., & Grassi, G. (2014). The Autonomic Nervous System and Hypertension. CIRCULATION RESEARCH, 114 (11), 1804-1814. https://doi.org/10. 1161/ CIRCRESAHA. 114. 302524.

[7] Kira, K., & Rendell, L. A. (1992). The feature selection problem: Traditional methods and a new algorithm. In *Proceedings of the Tenth National Conference on Artificial Intelligence (AAAI-92)* (pp. 129-134).

[8] Thambiraj, G., Gandhi, U., Mangalanathan, U., Jose, V. J. M., & Anand, M. (2020). Investigation on the effect of Womersley number, ECG and PPG features for cuff less blood pressure estimation using machine learning. Biomedical Signal Processing and Control, 60, 101942. https://doi.org/https:// doi. org/ 10.1016/j.bspc.2020.101942.

[9] Song, M. S., & Lee, S. B. (2026). Interpretable machine learning for hypertension detection using photoplethysmography (PPG) signals and their derivatives. Biomedical Signal Processing and Control, 113, Article 109194. https://doi.org/10.1016/j.bspc.2025.109194.

[10] Slapnicar, G., Mlakar, N., & Lustrek, M. (2019). Blood Pressure Estimation from Photoplethysmogram Using a Spectro-Temporal Deep Neural Network. SENSORS, 19(15), Article 3420. https://doi.org/10.3390/s19153420.

[11] Li, Y. H., Harfiya, L. N., Purwandari, K., & Lin, Y. D. (2020). Real-Time Cuffless Continuous Blood Pressure Estimation Using Deep Learning Model. SENSORS, 20(19), Article 5606. https:// doi.org/10.3390/s20195606.

[12] Flórez-López, R. (2002). Reviewing RELIEF and its extensions:: A new approach for estimating attributes considering high-correlated features 2002 IEEE INTERNATIONAL CONFERENCE ON DATA MINING, PROCEEDINGS.

[13] Basu, S., Kumbier, K., Brown, J. B., & Yu, B. (2018). Iterative random forests to discover predictive and stable high-order interactions. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 115(8), 1943-1948. https://doi.org/ 10.1073/ pnas. 1711236115.

[14] Xue, J. K., & Shen, B. (2020). A novel swarm intelligence optimization approach: sparrow search algorithm. SYSTEMS SCIENCE & CONTROL ENGINEERING, 8(1), 22-34. https:// doi.org/10.1080/21642583.2019.1708830.

[15] A.E.W. Johnson, T.J. Pollard, L. Shen, Data Descriptor: MIMIC-III, a freely accessible critical care database, Sci. Data 35 (2016).

[16] Alian, A. A., & Shelley, K. H. (2014). Photoplethysmography [Journal Article]. Best practice & research. Clinical anaesthesiology, 28(4), 395-406. https://doi.org/ 10.1016/j. bpa. 2014. 08.006.

[17] Khawaja, R. A., Qureshi, R., Mansure, A. H., & Yahya, M. E. (2010). Validation of Datascope Accutorr Plus using British Hypertension Society (BHS) and Association for the Advancement of Medical Instrumentation (AAMI) protocol guidelines [Journal Article]. Journal of the Saudi Heart Association, 22(1), 1-5. https:// doi.org/10. 1016/j.jsha. 2010. 03. 001.