Blood Pressure Estimation Research Articles

Remote Blood Pressure Estimation via the Spatiotemporal Mapping of Facial Videos

Blood pressure (BP) monitoring is vital in daily healthcare, especially for cardiovascular diseases. However, BP values are mainly acquired through a contact-sensing method, which is inconvenient and unfriendly for BP monitoring. This paper proposes an efficient end-to-end network for estimating BP values from a facial video to achieve remote BP estimation in daily life. The network first derives a spatiotemporal map of a facial video. Then, it regresses the BP ranges with a designed blood pressure classifier and simultaneously calculates the specific value with a blood pressure calculator in each BP range based on the spatiotemporal map. In addition, an innovative oversampling training strategy was developed to handle the problem of unbalanced data distribution. Finally, we trained the proposed blood pressure estimation network on a private dataset, MPM-BP, and tested it on a popular public dataset, MMSE-HR. As a result, the proposed network achieved a mean absolute error (MAE) and root mean square error (RMSE) of 12.35 mmHg and 16.55 mmHg on systolic BP estimations, and those for diastolic BP were 9.54 mmHg and 12.22 mmHg, which were better than the values obtained in recent works. It can be concluded that the proposed method has excellent potential for camera-based BP monitoring in the indoor scenarios in the real world.

Intensive care photoplethysmogram datasets and machine-learning for blood pressure estimation: Generalization not guarantied

The large MIMIC waveform dataset, sourced from intensive care units, has been used extensively for the development of Photoplethysmography (PPG) based blood pressure (BP) estimation algorithms. Yet, because the data comes from patients in severe conditions-often under the effect of drugs-it is regularly noted that the relationship between BP and PPG signal characteristics may be anomalous, a claim that we investigate here. A sample of 12,000 records from the MIMIC waveform dataset was stacked up against the 219 records of the PPG-BP dataset, an alternative public dataset obtained under controlled experimental conditions. The distribution of systolic and diastolic BP data and 31 PPG pulse morphological features was first compared between datasets. Then, the correlation between features and BP, as well as between the features themselves, was analysed. Finally, regression models were trained for each dataset and validated against the other. Statistical analysis showed significant differences between the datasets in diastolic BP and in 20 out of 31 features when adjusting for heart rate differences. The eight features showing the highest rank correlation to systolic BP in PPG-BP all displayed muted correlation levels in MIMIC. Regression tests showed twice higher baseline predictive power with PPG-BP than with MIMIC. Cross-dataset regression displayed a practically complete loss of predictive power for all models. The differences between the MIMIC and PPG-BP dataset exposed in this study suggest that BP estimation models based on the MIMIC dataset have reduced predictive power on the general population.

Continuous blood pressure monitoring by photoplethysmography - signal preprocessing requirements based on blood flow modelling

Objective. The aim of the study is to investigate the effect of the signal sampling frequency and low-pass filtering on the accuracy of the localisation of the fiducial points of the photoplethysmographic signal (PPG), and thus on the estimation of the blood pressure (i.e. the accuracy of the estimation). Approach. Statistical analysis was performed on 3,799 data samples taken from a publicly available database. Four PPG fiducial points of each sample signal were examined in the study. Main results. Simulation suggests that for noise-free data, cubic spline interpolation causes the sampling frequency (in the considered range of 62.5–500 Hz) to have only limited influence on localisation of the fiducial point. Better results were obtained for the pulse transit time (PTT) than pulse arrival time (PAT) approach. The acceptable filter band depends on the selected fiducial point and PAT or PTT approach. The best results were obtained for the tangent fiducial point. Significance. The presented results make it possible to estimate the minimum requirements for the sampling frequency and filtering of the PPG signal in order to obtain a reliable estimation of blood pressure.

Remote Health Monitoring System for the Estimation of Blood Pressure, Heart Rate, and Blood Oxygen Saturation Level

This article presents the design and implementation of an Internet of Things (IoT)-based remote health monitoring system for the estimation of blood pressure (BP), heart rate (HR), and blood oxygen saturation levels (SpO2). Our designed sensor can remotely monitor BP, HR, and SpO2. Our device collects, evaluates, predicts, and reads health data and then stores it on a remote platform named “ThinkSpeak,” which forms an IoT platform, with a 0.91 organic light-emitting diodes (OLEDs) screen display for viewing numerical health readings locally. We used a biomedical sensor device with an embedded signal condition unit, and a single photoplethysmography (PPG) signal was employed to derive and measure the PPG signal. A computer-based algorithm was generated, which factored in selected beneficial parameters measured from a single bio-inspired PPG signal. The measured PPG signal was used to estimate the individual user’s BP (both systolic and diastolic values), HR, and SpO2. An automatic multiscale-based peak (AMBP) detection algorithm was developed to obtain the maximum peak of the PPG signal. Furthermore, the developed sensor was benchmarked against two standard commercially available measurement devices: a Contec ambulatory BP sensor and a Braun pulse oximeter monitor. Our developed sensor is worn as a ring sensor and is interfaced with an Arduino 1010 WIFI MKR for remote health monitoring. Our estimated BP, HR, and SpO2 values were remotely monitored and a graphical representation was constructed.

Noninvasive continuous blood pressure estimation with fewer parameters based on RA-ReliefF feature selection and MPGA-BPN models

Photoplethysmography (PPG) and electrocardiogram (ECG) are closely related to the cardiovascular system and have substantial waveform parameters associated with blood pressure (BP), so extracting their waveform parameters to estimate BP has been widely studied. However, the more parameters, the more complex the model is, and the more factors affect the estimation results, which further affects the application of the BP estimation model to related equipment. Greatly reducing the number of parameters without reducing the accuracy of BP prediction, so as to improve the availability, applicability and robustness of BP estimation model, is the key for the real application of this method. In this study, we used relativity analysis (RA) to fill the gap of ReliefF algorithm, so the same number of parameter dimensions selected by ReliefF contain more signal information. Furthermore, a BP prediction model based on multi-population genetic algorithm (MPGA) and back propagation network (BPN) was constructed. Using waveform parameters of PPG and ECG signals with a duration of 20 s and the measured BP as a set of data, we extracted 522 and 23,608 sets of data from 25 healthy subjects and the UCI_BP dataset respectively. Three, five, seven and all parameters were selected by ReliefF for training in MPGA-BPN model and the results were compared. The comparison shows that five-parameters dataset can performs as all- parameters in both healthy individuals and UCI_BP, which confirms the validity of the feature selection in this study and lays the roots for BP estimation with fewer parameters and high accuracy.

Upper-Arm Photoplethysmographic Sensor with One-Time Calibration for Long-Term Blood Pressure Monitoring.

Wearable cuffless photoplethysmographic blood pressure monitors have garnered widespread attention in recent years; however, the long-term performance values of these devices are questionable. Most cuffless blood pressure monitors require initial baseline calibration and regular recalibrations with a cuffed blood pressure monitor to ensure accurate blood pressure estimation, and their estimation accuracy may vary over time if left uncalibrated. Therefore, this study assessed the accuracy and long-term performance of an upper-arm, cuffless photoplethysmographic blood pressure monitor according to the ISO 81060-2 standard. This device was based on a nonlinear machine-learning model architecture with a fine-tuning optimized method. The blood pressure measurement protocol followed a validation procedure according to the standard, with an additional four weekly blood pressure measurements over a 1-month period, to assess the long-term performance values of the upper-arm, cuffless photoplethysmographic blood pressure monitor. The results showed that the photoplethysmographic signals obtained from the upper arm had better qualities when compared with those measured from the wrist. When compared with the cuffed blood pressure monitor, the means ± standard deviations of the difference in BP at week 1 (baseline) were -1.36 ± 7.24 and -2.11 ± 5.71 mmHg for systolic and diastolic blood pressure, respectively, which met the first criterion of ≤5 ± ≤8.0 mmHg and met the second criterion of a systolic blood pressure ≤ 6.89 mmHg and a diastolic blood pressure ≤ 6.84 mmHg. The differences in the uncalibrated blood pressure values between the test and reference blood pressure monitors measured from week 2 to week 5 remained stable and met both criteria 1 and 2 of the ISO 81060-2 standard. The upper-arm, cuffless photoplethysmographic blood pressure monitor in this study generated high-quality photoplethysmographic signals with satisfactory accuracy at both initial calibration and 1-month follow-ups. This device could be a convenient and practical tool to continuously measure blood pressure over long periods of time.

Blood pressure stratification using photoplethysmography and light gradient boosting machine.

Introduction: Globally, hypertension (HT) is a substantial risk factor for cardiovascular disease and mortality; hence, rapid identification and treatment of HT is crucial. In this study, we tested the light gradient boosting machine (LightGBM) machine learning method for blood pressure stratification based on photoplethysmography (PPG), which is used in most wearable devices. Methods: We used 121 records of PPG and arterial blood pressure (ABP) signals from the Medical Information Mart for Intensive Care III public database. PPG, velocity plethysmography, and acceleration plethysmography were used to estimate blood pressure; the ABP signals were used to determine the blood pressure stratification categories. Seven feature sets were established and used to train the Optuna-tuned LightGBM model. Three trials compared normotension (NT) vs. prehypertension (PHT), NT vs. HT, and NT + PHT vs. HT. Results: The F1 scores for these three classification trials were 90.18%, 97.51%, and 92.77%, respectively. The results showed that combining multiple features from PPG and its derivative led to a more accurate classification of HT classes than using features from only the PPG signal. Discussion: The proposed method showed high accuracy in stratifying HT risks, providing a noninvasive, rapid, and robust method for the early detection of HT, with promising applications in the field of wearable cuffless blood pressure measurement.

Photoplethysmography Signal Wavelet Enhancement and Novel Features Selection for Non-Invasive Cuff-Less Blood Pressure Monitoring.

In this paper, new features relevant to blood pressure (BP) estimation using photoplethysmography (PPG) are presented. A total of 195 features, including the proposed ones and those already known in the literature, have been calculated on a set composed of 50,000 pulses from 1080 different patients. Three feature selection methods, namely Correlation-based Feature Selection (CFS), RReliefF and Minimum Redundancy Maximum Relevance (MRMR), have then been applied to identify the most significant features for BP estimation. Some of these features have been extracted through a novel PPG signal enhancement method based on the use of the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the enhanced signal leads to a reliable identification of the characteristic points of the PPG signal (e.g., systolic, diastolic and dicrotic notch points) by simple means, obtaining results comparable with those from purposely defined algorithms. For systolic points, mean and std of errors computed as the difference between the locations obtained using a purposely defined already known algorithm and those using the MODWT enhancement are, respectively, 0.0097 s and 0.0202 s; for diastolic points they are, respectively, 0.0441 s and 0.0486 s; for dicrotic notch points they are 0.0458 s and 0.0896 s. Hence, this study leads to the selection of several new features from the MODWT enhanced signal on every single pulse extracted from PPG signals, in addition to features already known in the literature. These features can be employed to train machine learning (ML) models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) in a non-invasive way, which is suitable for telemedicine health-care monitoring.

Combining Gaussian Process with Hybrid Optimal Feature Decision in Cuffless Blood Pressure Estimation.

Noninvasive blood pressure estimation is crucial for cardiovascular and hypertension patients. Cuffless-based blood pressure estimation has received much attention recently for continuous blood pressure monitoring. This paper proposes a new methodology that combines the Gaussian process with hybrid optimal feature decision (HOFD) in cuffless blood pressure estimation. First, we can choose one of the feature selection methods: robust neighbor component analysis (RNCA), minimum redundancy, maximum relevance (MRMR), and F-test, based on the proposed hybrid optimal feature decision. After that, a filter-based RNCA algorithm uses the training dataset to obtain weighted functions by minimizing the loss function. Next, we combine the Gaussian process (GP) algorithm as the evaluation criteria, which is used to determine the best feature subset. Hence, combining GP with HOFD leads to an effective feature selection process. The proposed combining Gaussian process with the RNCA algorithm shows that the root mean square errors (RMSEs) for the SBP (10.75 mmHg) and DBP (8.02 mmHg) are lower than those of the conventional algorithms. The experimental results represent that the proposed algorithm is very effective.

The pre-ejection period is a highly stress dependent parameter of paramount importance for pulse-wave-velocity based applications.

The pulse-wave-velocity, is used for indirect, cuff-less, continuous blood pressure estimation. It is commonly detected by measuring the time delay between a defined point in an ECG and the arrival of the peripheral pulse wave (e.g., oxygen saturation sensor). The period between electrical stimulation of the heart (ECG) and actual blood ejection from the heart is called the pre-ejection period (PEP). This study aims at characterizing the PEP under mental and physical stress with focus on its relations to other cardiovascular parameters such as heart rate and importance for blood pressure (BP) estimation. We measured the PEP in 71 young adults at rest, under mental (TSST) and physical stress (ergometer) via impedance-cardiography. The PEP is highly dependent on mental and physical load. It is strongly correlated with indicators of sympathetic strain (p < 0.001). At rest (mean 104.5 ms), the PEP shows a high interindividual variability but small intraindividual variability. Mental stress decreases the PEP by 16% (mean 90.0 ms) while physical stress halves PEP (mean 53.9 ms). The PEP does correlate differently with heart rate under differing circumstances (rest: R 2 0.06, mental stress: R 2 0.29, physical stress: R 2 0.65). Subsequently, using PEP and heart rate enables the discrimination of rest, mental and physical strain with a positive predictive value of 93%. The PEP is a cardiovascular parameter with large interindividual variability at rest and subject-depended dynamic under load which is of great importance for ECG-based pulse-wave-velocity (PWV) determination. Considering its variability and large impact on the pulse arrival time, PEP is a crucial factor in PWV based BP estimation.

Factors of QTc interval prolongation in patients with arterial hypertension

The relevance of determining the factors of the prolonged QTc interval development in patients with arterial hypertension is high, as it solves the issue of preliminary diagnosis among patients from risk groups and allows adjusting the diagnostic and therapeutic tactics of managing such patients. Aim: to determine factors for the development of prolonged average daily QTc interval in patients with arterial hypertension. Materials and methods. The study included 195 patients with arterial hypertension, who were divided into 2 subgroups: 175 individuals with a normal average daily duration of the QTc interval and 20 patients with a prolonged average daily duration of the QTc interval. All the patients underwent 24-hour heart rate monitoring and determination of blood pressure variability. At the stage of involving patients in the study, the following were performed: anamnesis collection and determination of physical data (estimation of systolic and diastolic blood pressure). Results. The study has revealed risk factors of daily QTc interval prolongation in hypertensive patients with further developing a prognostic model with high sensitivity (88.9 %) and specificity (70.6 %). Prolongation of QTc in patients with hypertension was associated with such parameters: male sex (OR = 4.292 [95.0 % CI 1.337–13.779], Р = 0.014), body mass index (OR = 1.107 [95.0 % CI 0.987–1.243], Р = 0.083), mean 24-hour diastolic (OR = 1.076 [95.0 % CI 1.017–1.139], Р = 0.011) and pulse blood pressure (OR = 1.075 [95.0 % CІ 1.012–1.141], Р = 0.018), time since hypertension was diagnosed (OR = 1.093 [95.0 % CI 0.997–1.197], Р = 0.057) and office pulse blood pressure (OR = 0.948 [95.0 % CІ 0.900–0.999], Р = 0.046). Conclusions. The study has determined risk factors of developing a prolonged average daily QTc interval in patients with hypertension, which will allow correcting diagnostic and treatment tactics of management for patients from risk groups.

Examining the Individual Response to a Low-Sodium Diet in Patients with Hypertension: Protocol for a Pilot Randomized Controlled Trial.

Excessive dietary sodium intake is an independent risk factor for hypertension and cardiovascular disease (CVD). Despite the large body of evidence concerning the effects of dietary interventions on blood pressure (BP) and CVD outcomes, trials have often reported low adherence to decreased sodium intake, likely due in part to heterogeneous BP responses. To address the challenges, recent clinical findings suggested a precise and personalized dietary approach that seeks to deliver more preventive and practical dietary advice than the "one-size-fits-all" guidelines and weighs the personal risk of developing specific diseases. The purpose of this pilot randomized controlled trial was to test the feasibility and preliminary efficacy of integrating the use of mobile technology and metabolomics with a low-sodium diet intervention in patients with hypertension to develop personalized low-sodium diet programs. Additionally, the study will examine the associations of urine metabolites with urinary sodium levels and BP control based on the hypothesis that targeted urine metabolites. In this report, we describe the design and protocol of the pilot trial. A total of 40 patients with hypertension will be randomly assigned to either a 8-week low-sodium diet group (n=20) or a standard care group (n=20). Each week, intervention participants went through individual sessions with an interventionist via videoconferencing to discuss low-sodium diet regimens, patients' food choices, and BP tracks on mobile apps. The control group followed their usual care for hypertension management. All participants in both groups monitored diet and BP using mobile apps for 8 weeks. A 24-hour urinary sodium excretion for the estimation of dietary sodium intake, systolic, and diastolic BPs were measured at the baseline and at 8 weeks. The primary outcomes of this study include the feasibility of conducting a randomized controlled trial (RCT) by reporting recruitment, retention, and completion statistics. The preliminary effects of intervention will be tested by a generalized estimating equation model. This pilot RCT study was approved by the institutional review board at the University of Texas Health San Antonio in January 2021. The first participant was enrolled in April 2021, and currently, 26 participants were enrolled. All data collection is expected to conclude by March 2023, with data analysis and study results ready for reporting by December 2023. Findings from this pilot RCT will further guide the team in planning a future large-scale study. The findings of this proposed study will establish a comprehensive knowledge base for future research and development of personalized dietary interventions to promote adherence to dietary strategies and self-management of chronic disease using the Precision Health approach for millions of Americans who are struggling with uncontrolled hypertension. ClinicalTrials.gov NCT04764253; https://clinicaltrials.gov/ct2/show/NCT04764253. DERR1-10.2196/39058.

PulseDB: A large, cleaned dataset based on MIMIC-III and VitalDB for benchmarking cuff-less blood pressure estimation methods.

There has been a growing interest in developing cuff-less blood pressure (BP) estimation methods to enable continuous BP monitoring from electrocardiogram (ECG) and/or photoplethysmogram (PPG) signals. The majority of these methods have been evaluated using publicly-available datasets, however, there exist significant discrepancies across studies with respect to the size, the number of subjects, and the applied pre-processing steps for the data that is eventually used for training and testing the models. Such differences make conducting performance comparison across models largely unfair, and mask the generalization capability of various BP estimation methods. To fill this important gap, this paper presents "PulseDB," the largest cleaned dataset to date, for benchmarking BP estimation models that also fulfills the requirements of standardized testing protocols. PulseDB contains 1) 5,245,454 high-quality -s segments of ECG, PPG, and arterial BP (ABP) waveforms from 5,361 subjects retrieved from the MIMIC-III waveform database matched subset and the VitalDB database; 2) subjects' identification and demographic information, that can be utilized as additional input features to improve the performance of BP estimation models, or to evaluate the generalizability of the models to data from unseen subjects; and 3) positions of the characteristic points of the ECG/PPG signals, making PulseDB directly usable for training deep learning models with minimal data pre-processing. Additionally, using this dataset, we conduct the first study to provide insights about the performance gap between calibration-based and calibration-free testing approaches for evaluating generalizability of the BP estimation models. We expect PulseDB, as a user-friendly, large, comprehensive and multi-functional dataset, to be used as a reliable source for the evaluation of cuff-less BP estimation methods.

Neural Network Model Combination for Video-Based Blood Pressure Estimation: New Approach and Evaluation.

One of the most effective vital signs of health conditions is blood pressure. It has such an impact that changes your state from completely relaxed to extremely unpleasant, which makes the task of blood pressure monitoring a main procedure that almost everyone undergoes whenever there is something wrong or suspicious with his/her health condition. The most popular and accurate ways to measure blood pressure are cuff-based, inconvenient, and pricey, but on the bright side, many experimental studies prove that changes in the color intensities of the RGB channels represent variation in the blood that flows beneath the skin, which is strongly related to blood pressure; hence, we present a novel approach to blood pressure estimation based on the analysis of human face video using hybrid deep learning models. We deeply analyzed proposed approaches and methods to develop combinations of state-of-the-art models that were validated by their testing results on the Vision for Vitals (V4V) dataset compared to the performance of other available proposed models. Additionally, we came up with a new metric to evaluate the performance of our models using Pearson's correlation coefficient between the predicted blood pressure of the subjects and their respiratory rate at each minute, which is provided by our own dataset that includes 60 videos of operators working on personal computers for almost 20 min in each video. Our method provides a cuff-less, fast, and comfortable way to estimate blood pressure with no need for any equipment except the camera of your smartphone.

Blood pressure monitoring during anesthesia induction using PPG morphology features and machine learning.

Blood pressure (BP) is a crucial biomarker giving valuable information regarding cardiovascular diseases but requires accurate continuous monitoring to maximize its value. In the effort of developing non-invasive, non-occlusive and continuous BP monitoring devices, photoplethysmography (PPG) has recently gained interest. Researchers have attempted to estimate BP based on the analysis of PPG waveform morphology, with promising results, yet often validated on a small number of subjects with moderate BP variations. This work presents an accurate BP estimator based on PPG morphology features. The method first uses a clinically-validated algorithm (oBPM®) to perform signal preprocessing and extraction of physiological features. A subset of features that best reflects BP changes is automatically identified by Lasso regression, and a feature relevance analysis is conducted. Three machine learning (ML) methods are then investigated to translate this subset of features into systolic BP (SBP) and diastolic BP (DBP) estimates; namely Lasso regression, support vector regression and Gaussian process regression. The accuracy of absolute BP estimates and trending ability are evaluated. Such an approach considerably improves the performance for SBP estimation over previous oBPM® technology, with a reduction in the standard deviation of the error of over 20%. Furthermore, rapid BP changes assessed by the PPG-based approach demonstrates concordance rate over 99% with the invasive reference. Altogether, the results confirm that PPG morphology features can be combined with ML methods to accurately track BP variations generated during anesthesia induction. They also reinforce the importance of adding a calibration measure to obtain an absolute BP estimate.

A Novel Triple-Bladder Cuff Method for Blood Pressure Estimation Based on Pulse Wave Dynamics in Brachial Artery: Theoretical and Experimental Analyses.

The objective of this study was to develop a novel triple-bladder cuff method for accurate and automated estimation of systolic (SBP) and diastolic (DBP) blood pressure and validate its reliability in animal experiments. The cuff is composed of three bladders each measured one-third the width of a conventional BP cuff, which are designed to measure oscillatory pulsation at the proximal, middle, and distal segments of the upper arm. This structure allows evaluation of the pulse wave propagation in the brachial artery under the cuff. SBP is estimated (SBPe) by detecting resumption of systolic arterial flow based on statistical similarity in oscillatory pulse traces between the proximal and distal segments. DBP is estimated (DBPe) based on the relation between pulse wave velocity and transmural pressure at diastole in the brachial artery. In 7 anesthetized goats, we compared SBPe and DBPe to reference SBP and DBP, respectively, measured by an intra-arterial catheter. BP was perturbed by infusing nitroprusside or noradrenaline. SBP correlated strongly with SBPe in each animal [mean coefficient of determination (R2) = 0.98 ± 0.01]. Mean ± standard deviation of errors between SBP and SBPe was 0.0 ± 4.9 mmHg. DBP correlated strongly with DBPe in each animal (R2 = 0.96 ± 0.03). Mean ± standard deviation of errors between DBP and DBPe was 0.0 ± 6.3 mmHg. This method estimates SBP and DBP with acceptable accuracy. Accurate and automated BP estimation by this method may potentially optimize antihypertensive treatment in patients with hypertension.

Features from the photoplethysmogram and the electrocardiogram for estimating changes in blood pressure

There is a growing emphasis being placed on the potential for cuffless blood pressure (BP) estimation through modelling of morphological features from the photoplethysmogram (PPG) and electrocardiogram (ECG). However, the appropriate features and models to use remain unclear. We investigated the best features available from the PPG and ECG for BP estimation using both linear and non-linear machine learning models. We conducted a clinical study in which changes in BP (DeltaBP) were induced by an infusion of phenylephrine in 30 healthy volunteers (53.8% female, 28.0 (9.0) years old). We extracted a large and diverse set of features from both the PPG and the ECG and assessed their individual importance for estimating DeltaBP through Shapley additive explanation values and a ranking coefficient. We trained, tuned, and evaluated linear (ordinary least squares, OLS) and non-linear (random forest, RF) machine learning models to estimate DeltaBP in a nested leave-one-subject-out cross-validation framework. We reported the results as correlation coefficient (rho _p), root mean squared error (RMSE), and mean absolute error (MAE). The non-linear RF model significantly (p<0.05) outperformed the linear OLS model using both the PPG and the ECG signals across all performance metrics. Estimating DeltaSBP using the PPG alone (rho _p = 0.86 (0.23), RMSE = 5.66 (4.76) mmHg, MAE = 4.86 (4.29) mmHg) performed significantly better than using the ECG alone (rho _p = 0.69 (0.45), RMSE = 6.79 (4.76) mmHg, MAE = 5.28 (4.57) mmHg), all p < 0.001. The highest ranking features from the PPG largely modelled increasing reflected wave interference driven by changes in arterial stiffness. This finding was supported by changes observed in the PPG waveform in response to the phenylephrine infusion. However, a large number of features were required for accurate BP estimation, highlighting the high complexity of the problem. We conclude that the PPG alone may be further explored as a potential single source, cuffless, blood pressure estimator. The use of the ECG alone is not justified. Non-linear models may perform better as they are able to incorporate interactions between feature values and demographics. However, demographics may not adequately account for the unique and individualised relationship between the extracted features and BP.

Cuffless Blood Pressure Estimation with Confidence Intervals using Hybrid Feature Selection and Decision Based on Gaussian Process

Cuffless blood pressure (BP) monitoring is crucial for patients with cardiovascular disease and hypertension. However, conventional BP monitors provide only single-point estimates without confidence intervals. Therefore, the statistical variability in the estimates is indistinguishable from the intrinsic variability caused by physiological processes. This study introduced a novel method for improving the reliability of BP and confidence intervals (CIs) estimations using a hybrid feature selection and decision method based on a Gaussian process. F-test and robust neighbor component analysis were applied as feature selection methods for obtaining a set of highly weighted features to estimate accurate BP and CIs. Akaike’s information criterion algorithm was used to select the best feature subset. The performance of the proposed algorithm was confirmed through experiments. Comparisons with conventional algorithms indicated that the proposed algorithm provided the most accurate BP and CIs estimates. To the best of the authors’ knowledge, the proposed method is currently the only one that provides highly reliable BP and CIs estimates. Therefore, the proposed algorithm may be robust for concurrently estimating BP and CIs.

EVALUATION OF AGREEMENT BETWEEN OSCILLOMETRIC AND DIRECT BLOOD PRESSURE MEASUREMENTS IN ANESTHETIZED TIGERS (PANTHERA TIGRIS).

Noninvasive blood pressure measurement is commonly performed with oscillometry; however, this technique provides clinically helpful information only if it is representative of the gold standard. Agreement between direct and oscillometric blood pressure measurements were performed in 14 anesthetized, captive tigers (Panthera tigris). A cuff, placed around the tail base and connected to a multiparameter monitor, was used to measure arterial blood pressure oscillometrically and provided systolic, mean, and diastolic pressures. At the same time, direct blood pressures were obtained from a dorsal pedal arterial catheter, and the oscillometric and direct readings were considered paired data points. Agreement between the two methods was evaluated by Bland-Altman plots. All animals completed the study and provided 196 paired data points. The bias (mm Hg) for systolic, mean, and diastolic arterial pressures was -3.7, -0.8, and -1.6, respectively. Limits of agreement (mm Hg) for systolic, mean, and diastolic arterial pressures were -31 to 24, -29 to 27, and -29 to 26, respectively. Oscillometry provided an acceptable amount of readings within 10 and 20 mm Hg of the gold standard. The oscillometric technique provided reasonable agreement with direct measurements. Therefore, in the conditions used in this study, oscillometric blood pressure measured via the ventral coccygeal artery provided reasonable estimates of invasive blood pressure in anesthetized tigers.

Noncontact Blood Pressure Estimation Using BP-Related Cardiovascular Knowledge: An Uncalibrated Method Based on Consumer-Level Camera

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective</i> : The tiny change of skin color, caused by heartbeat, can be captured with consumer-level cameras by using the imaging photoplethysmography (iPPG) technique, offering a non-contact way of extracting pulse signals. The pulse signals have been demonstrated to contain information of human physiological characteristics and have been used for blood pressure (BP) estimation in recent years. According to BP-related cardiovascular knowledge, this paper presents a new method for BP estimation based on the iPPG pulse signals, featured by incorporating cardiovascular characteristics including heart rate (HR), stroke volume (SV), elasticity of vessel walls (EVW), and peripheral vascular resistance (PVR). Correlations between the systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure (PP), and the cardiovascular characteristics are extracted, which facilitates the selection of pulse features consistent with BP properties. Based on the selected features, two Bayesian neural network (BNN) models are constructed for the estimation of SBP and DBP respectively, where the machine learning (ML) uncertainty of the estimation is also evaluated. This method is uncalibrated that means it can work without additional information except for the videos from the camera. The proposed method has been tested on 220 patients with a history of cardiovascular diseases. Errors of the BP estimation are 9±13 (MAE±STD) mmHg for SBP, 7±10 (MAE±STD) mmHg for DBP, and the ML uncertainty of the estimation indicates reliability of the proposed method.