Estimation of cardiac output and systemic vascular resistance using a multivariate regression model with features selected from the finger photoplethysmogram and routine cardiovascular measurements

BackgroundPhotoplethysmography (PPG) is a proven way to measure heart rate (HR). This technology is already available in smartphones, which allows measuring HR only by using the smartphone. Given the widespread availability of smartphones, this creates a scalable way to enable mobile HR monitoring. An essential precondition is that these technologies are as reliable and accurate as the current clinical (gold) standards. At this moment, there is no consensus on a gold standard method for the validation of HR apps. This results in different validation processes that do not always reflect the veracious outcome of comparison.ObjectiveThe aim of this paper was to investigate and describe the necessary elements in validating and comparing HR apps versus standard technology.MethodsThe FibriCheck (Qompium) app was used in two separate prospective nonrandomized studies. In the first study, the HR of the FibriCheck app was consecutively compared with 2 different Food and Drug Administration (FDA)-cleared HR devices: the Nonin oximeter and the AliveCor Mobile ECG. In the second study, a next step in validation was performed by comparing the beat-to-beat intervals of the FibriCheck app to a synchronized ECG recording.ResultsIn the first study, the HR (BPM, beats per minute) of 88 random subjects consecutively measured with the 3 devices showed a correlation coefficient of .834 between FibriCheck and Nonin, .88 between FibriCheck and AliveCor, and .897 between Nonin and AliveCor. A single way analysis of variance (ANOVA; P=.61 was executed to test the hypothesis that there were no significant differences between the HRs as measured by the 3 devices. In the second study, 20,298 (ms) R-R intervals (RRI)–peak-to-peak intervals (PPI) from 229 subjects were analyzed. This resulted in a positive correlation (rs=.993, root mean square deviation [RMSE]=23.04 ms, and normalized root mean square error [NRMSE]=0.012) between the PPI from FibriCheck and the RRI from the wearable ECG. There was no significant difference (P=.92) between these intervals.ConclusionsOur findings suggest that the most suitable method for the validation of an HR app is a simultaneous measurement of the HR by the smartphone app and an ECG system, compared on the basis of beat-to-beat analysis. This approach could lead to more correct assessments of the accuracy of HR apps.

show abstract

“…Additionally, it can also be used to estimate cardiac output [8]. PPG used as signal to measure HR is described as the pulse signal.…”

Section: Introductionmentioning

confidence: 99%

Clinical Validation of Heart Rate Apps: Mixed-Methods Evaluation Study

Vandenberk¹,

Stans²,

Mortelmans³

et al. 2017

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

show abstract

“…Heart rate variability (HRV) is a proven indicator to reflect ANS. PPG variability (PPGV) has also been shown to be associated with sympathetic vasomotor activities (Nitzan et al 1998, Middleton et al 2011, Lee et al 2013. Thus, HRV and PPGV were included as features in the model where PPGV was derived by calculating PPG amplitude variability.…”

Section: Frequency Spectrummentioning

confidence: 99%

Systolic blood pressure estimation using PPG and ECG during physical exercise

Sun¹,

Bezemer²,

Long³

et al. 2016

Physiol. Meas.

View full text Add to dashboard Cite

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.

show abstract

“…Blood oxygen saturation (Kyriacou et al, 2002), pulse number (Selvaraj et al, 2008), respiratory rate (Johansson, 2003) and blood pressure (Binns et al, 1995) were the major PPG parameters studied in research. Furthermore, PPG could contain some other significant cardiovascular features, such as arterial stiffness (Pilt et al, 2013), cardiac output and systemic vascular resistance (Lee et al, 2013). Until to now many properties of PPG were analyzed, but attributes and physiological characteristics of PPG signal not fully discovered yet (Li et al, 2014).…”

Section: Introductionmentioning

confidence: 99%

Biological Assessments by Innovative Use of Multi-Wavelength Photoplethysmographic Signals Time Differences

Vahdani-Ma¹

2015

J. of Applied Sciences

View full text Add to dashboard Cite

A new biological assessment technique was presented by novel use of photoplethysmography (PPG) signals. Most previous PPG signal analysis techniques were based on morphological and frequency characteristics of PPG. In the present study, time differences (TDs) between separate pairs of PPG signal (with different wavelengths) were used as the novel aspect of PPG technology. A computer connectable multi-channel PPG device and a MATLAB ® based program were used to record transmission mode PPG signals with wavelengths of 940 nm (IR), 660 nm (red), 520 nm (green) and 450 nm (blue). The TDs between three different pairs of the signals were obtained during a breath holding experiment (as biology challenge test) to investigate and compare their Time Difference (TD) variations within the tests. The experiments were performed with the participation of 25 healthy subjects. Results showed that TDs of PPG signal pairs had significant variations (p<0.05) in reply to biological challenges. It was concluded that TD variations can be indicative of very important biological information. A good attribute of PPG TD method was its independence from signal amplitude and avoiding from corresponding disadvantage.

show abstract

Estimation of cardiac output and systemic vascular resistance using a multivariate regression model with features selected from the finger photoplethysmogram and routine cardiovascular measurements

Cited by 23 publications

References 25 publications

Clinical Validation of Heart Rate Apps: Mixed-Methods Evaluation Study

Clinical Validation of Heart Rate Apps: Mixed-Methods Evaluation Study

Systolic blood pressure estimation using PPG and ECG during physical exercise

Biological Assessments by Innovative Use of Multi-Wavelength Photoplethysmographic Signals Time Differences

Contact Info

Product

Resources

About