Systolic blood pressure estimation using PPG and ECG during physical exercise

S. Sun, R. Bezemer, X. Long, J. Muehlsteff, R.M. Aarts

Research output: Contribution to journalArticleAcademicpeer-review

9 Citations (Scopus)
1 Downloads (Pure)

Abstract

In this work, a model to estimate systolic blood pressure (SBP) using photoplethysmography (PPG) and electrocardiography (ECG) is proposed. Data from 19 subjects doing a 40 min exercise was analyzed. Reference SBP was measured at the finger based on the volume-clamp principle. PPG signals were measured at the finger and forehead. After an initialization process for each subject at rest, the model estimated SBP every 30 s for the whole period of exercise. In order to build this model, 18 features were extracted from PPG signals by means of its waveform, first derivative, second derivative, and frequency spectrum. In addition, pulse arrival time (PAT) was derived as a feature from the combination of PPG and ECG. After evaluating four regression models, we chose multiple linear regression (MLR) to combine all derived features to estimate SBP. The contribution of each feature was quantified using its normalized weight in the MLR. To evaluate the performance of the model, we used a leave-one-subject-out cross validation. With the aim of exploring the potential of the model, we investigated the influences of the inclusion of PAT, regression models, measurement sites (finger and forehead), and posture change (lying, sitting, and standing). The results show that the inclusion of PAT reduced the standard deviation (SD) of the difference from 14.07 to 13.52 mmHg. There was no significant difference in the estimation performance between the model using finger- and forehead-derived PPG signals. Separate models are necessary for different postures. The optimized model using finger-derived PPG signals during physical exercise had a performance with a mean difference of 0.43 mmHg, an SD of difference of 13.52 mmHg, and median correlation coefficients of 0.86. Furthermore, we identified two groups of features that contributed more to SBP estimation compared to other features. One group consists of our proposed features depicting beat morphology. The other comprises existing features depicting the dicrotic notch. The present work demonstrates promising results of the SBP estimation model during physical exercise.
Original languageEnglish
Pages (from-to)2154-2169
Number of pages16
JournalPhysiological Measurement
Volume37
Issue number12
DOIs
Publication statusPublished - 14 Nov 2016

Fingerprint

Photoplethysmography
Blood pressure
Electrocardiography
Exercise
Blood Pressure
Fingers
Forehead
Pulse
Linear regression
Posture
Linear Models
Derivatives
Clamping devices

Keywords

  • photoplethysmography
  • systolic blood pressure
  • multiple linear regression
  • pulse arrival time

Cite this

Sun, S. ; Bezemer, R. ; Long, X. ; Muehlsteff, J. ; Aarts, R.M. / Systolic blood pressure estimation using PPG and ECG during physical exercise. In: Physiological Measurement. 2016 ; Vol. 37, No. 12. pp. 2154-2169.
@article{86dbd49e512a4a0f95348595cbf4fe9a,
title = "Systolic blood pressure estimation using PPG and ECG during physical exercise",
abstract = "In this work, a model to estimate systolic blood pressure (SBP) using photoplethysmography (PPG) and electrocardiography (ECG) is proposed. Data from 19 subjects doing a 40 min exercise was analyzed. Reference SBP was measured at the finger based on the volume-clamp principle. PPG signals were measured at the finger and forehead. After an initialization process for each subject at rest, the model estimated SBP every 30 s for the whole period of exercise. In order to build this model, 18 features were extracted from PPG signals by means of its waveform, first derivative, second derivative, and frequency spectrum. In addition, pulse arrival time (PAT) was derived as a feature from the combination of PPG and ECG. After evaluating four regression models, we chose multiple linear regression (MLR) to combine all derived features to estimate SBP. The contribution of each feature was quantified using its normalized weight in the MLR. To evaluate the performance of the model, we used a leave-one-subject-out cross validation. With the aim of exploring the potential of the model, we investigated the influences of the inclusion of PAT, regression models, measurement sites (finger and forehead), and posture change (lying, sitting, and standing). The results show that the inclusion of PAT reduced the standard deviation (SD) of the difference from 14.07 to 13.52 mmHg. There was no significant difference in the estimation performance between the model using finger- and forehead-derived PPG signals. Separate models are necessary for different postures. The optimized model using finger-derived PPG signals during physical exercise had a performance with a mean difference of 0.43 mmHg, an SD of difference of 13.52 mmHg, and median correlation coefficients of 0.86. Furthermore, we identified two groups of features that contributed more to SBP estimation compared to other features. One group consists of our proposed features depicting beat morphology. The other comprises existing features depicting the dicrotic notch. The present work demonstrates promising results of the SBP estimation model during physical exercise.",
keywords = "photoplethysmography, systolic blood pressure, multiple linear regression, pulse arrival time",
author = "S. Sun and R. Bezemer and X. Long and J. Muehlsteff and R.M. Aarts",
year = "2016",
month = "11",
day = "14",
doi = "10.1088/0967-3334/37/12/2154",
language = "English",
volume = "37",
pages = "2154--2169",
journal = "Physiological Measurement",
issn = "0967-3334",
publisher = "Institute of Physics",
number = "12",

}

Systolic blood pressure estimation using PPG and ECG during physical exercise. / Sun, S.; Bezemer, R.; Long, X.; Muehlsteff, J.; Aarts, R.M.

In: Physiological Measurement, Vol. 37, No. 12, 14.11.2016, p. 2154-2169.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Systolic blood pressure estimation using PPG and ECG during physical exercise

AU - Sun, S.

AU - Bezemer, R.

AU - Long, X.

AU - Muehlsteff, J.

AU - Aarts, R.M.

PY - 2016/11/14

Y1 - 2016/11/14

N2 - In this work, a model to estimate systolic blood pressure (SBP) using photoplethysmography (PPG) and electrocardiography (ECG) is proposed. Data from 19 subjects doing a 40 min exercise was analyzed. Reference SBP was measured at the finger based on the volume-clamp principle. PPG signals were measured at the finger and forehead. After an initialization process for each subject at rest, the model estimated SBP every 30 s for the whole period of exercise. In order to build this model, 18 features were extracted from PPG signals by means of its waveform, first derivative, second derivative, and frequency spectrum. In addition, pulse arrival time (PAT) was derived as a feature from the combination of PPG and ECG. After evaluating four regression models, we chose multiple linear regression (MLR) to combine all derived features to estimate SBP. The contribution of each feature was quantified using its normalized weight in the MLR. To evaluate the performance of the model, we used a leave-one-subject-out cross validation. With the aim of exploring the potential of the model, we investigated the influences of the inclusion of PAT, regression models, measurement sites (finger and forehead), and posture change (lying, sitting, and standing). The results show that the inclusion of PAT reduced the standard deviation (SD) of the difference from 14.07 to 13.52 mmHg. There was no significant difference in the estimation performance between the model using finger- and forehead-derived PPG signals. Separate models are necessary for different postures. The optimized model using finger-derived PPG signals during physical exercise had a performance with a mean difference of 0.43 mmHg, an SD of difference of 13.52 mmHg, and median correlation coefficients of 0.86. Furthermore, we identified two groups of features that contributed more to SBP estimation compared to other features. One group consists of our proposed features depicting beat morphology. The other comprises existing features depicting the dicrotic notch. The present work demonstrates promising results of the SBP estimation model during physical exercise.

AB - In this work, a model to estimate systolic blood pressure (SBP) using photoplethysmography (PPG) and electrocardiography (ECG) is proposed. Data from 19 subjects doing a 40 min exercise was analyzed. Reference SBP was measured at the finger based on the volume-clamp principle. PPG signals were measured at the finger and forehead. After an initialization process for each subject at rest, the model estimated SBP every 30 s for the whole period of exercise. In order to build this model, 18 features were extracted from PPG signals by means of its waveform, first derivative, second derivative, and frequency spectrum. In addition, pulse arrival time (PAT) was derived as a feature from the combination of PPG and ECG. After evaluating four regression models, we chose multiple linear regression (MLR) to combine all derived features to estimate SBP. The contribution of each feature was quantified using its normalized weight in the MLR. To evaluate the performance of the model, we used a leave-one-subject-out cross validation. With the aim of exploring the potential of the model, we investigated the influences of the inclusion of PAT, regression models, measurement sites (finger and forehead), and posture change (lying, sitting, and standing). The results show that the inclusion of PAT reduced the standard deviation (SD) of the difference from 14.07 to 13.52 mmHg. There was no significant difference in the estimation performance between the model using finger- and forehead-derived PPG signals. Separate models are necessary for different postures. The optimized model using finger-derived PPG signals during physical exercise had a performance with a mean difference of 0.43 mmHg, an SD of difference of 13.52 mmHg, and median correlation coefficients of 0.86. Furthermore, we identified two groups of features that contributed more to SBP estimation compared to other features. One group consists of our proposed features depicting beat morphology. The other comprises existing features depicting the dicrotic notch. The present work demonstrates promising results of the SBP estimation model during physical exercise.

KW - photoplethysmography

KW - systolic blood pressure

KW - multiple linear regression

KW - pulse arrival time

U2 - 10.1088/0967-3334/37/12/2154

DO - 10.1088/0967-3334/37/12/2154

M3 - Article

C2 - 27841157

VL - 37

SP - 2154

EP - 2169

JO - Physiological Measurement

JF - Physiological Measurement

SN - 0967-3334

IS - 12

ER -