Estimating pulse wave velocity from the radial pressure wave using machine learning algorithms

Association between estimated pulse wave velocity and in-hospital mortality of patients with acute kidney injury: a retrospective cohort analysis of the MIMIC-IV database

BACKGROUND

Estimated pulse wave velocity (ePWV) has been found to be an independent predictor of cardiovascular mortality and kidney injury, which can be estimated noninvasively. This study aimed to investigate the association between ePWV and in-hospital mortality in critically ill patients with acute kidney injury (AKI).

METHODS

This study included 5960 patients with AKI from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The low and high ePWV groups were compared using a Kaplan-Meier survival curve to evaluate the differences in survival status. Cox proportional hazards models were used to explore the association between ePWV and in-hospital mortality in critically ill patients with AKI. To further examine the dose-response relationship, we used a restricted cubic spline (RCS) model. Stratification analyses were conducted to investigate the effect of ePWV on hospital mortality across various subgroups.

RESULTS

Survival analysis indicated that patients with high ePWV had a lower survival rate than those with low ePWV. Following adjustment, high ePWV demonstrated a statistically significant association with an increased risk of in-hospital mortality among AKI patients (HR = 1.53, 95% CI = 1.36-1.71, p < 0.001). Analysis using the RCS model confirmed a linear increase in the risk of hospital mortality as the ePWV values increased (P for nonlinearity = 0.602).

CONCLUSIONS

A high ePWV was significantly associated with an increased risk of in-hospital mortality among patients with AKI. Furthermore, ePWV was an independent predictor of in-hospital mortality in critically ill patients with AKI.

Association between estimated pulse wave velocity and in-hospital and one-year mortality of patients with chronic kidney disease and atherosclerotic heart disease: a retrospective cohort analysis of the MIMIC-IV database

BACKGROUND

Carotid-femoral pulse wave velocity has been identified as an autonomous predictor of cardiovascular mortality and kidney injury. This important clinical parameter can be non-invasively estimated using the calculated pulse wave velocity (ePWV). The objective of this study was to examine the correlation between ePWV and in-hospital as well as one-year mortality among critically ill patients with chronic kidney disease (CKD) and atherosclerotic heart disease (ASHD).

METHODS

This study included a cohort of 1173 patients diagnosed with both CKD and ASHD, sourced from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The four groups divided into quartiles according to ePWV were compared using a Kaplan-Meier survival curve to assess variations in survival rates. Cox proportional hazards models were employed to analyze the correlation between ePWV and in-hospital as well as one-year mortality among critically ill patients with both CKD and ASHD. To further investigate the dose-response relationship, a restricted cubic splines (RCS) model was utilized. Additionally, stratification analyses were performed to examine the impact of ePWV on hospital and one-year mortality across different subgroups.

RESULTS

The survival analysis results revealed a negative correlation between higher ePWV and survival rate. After adjusting for confounding factors, higher ePWV level (ePWV > 11.90 m/s) exhibited a statistically significant association with an increased risk of both in-hospital and one-year mortality among patients diagnosed with both CKD and ASHD (HR = 4.72, 95% CI = 3.01-7.39, p < 0.001; HR = 2.04, 95% CI = 1.31-3.19, p = 0.002). The analysis incorporating an RCS model confirmed a linear escalation in the risk of both in-hospital and one-year mortality with rising ePWV values (P for nonlinearity = 0.619; P for nonlinearity = 0.267).

CONCLUSIONS

The ePWV may be a potential marker for the in-hospital and one-year mortality assessment of CKD with ASHD, and elevated ePWV was strongly correlated with an elevated mortality risk in patients diagnosed with both CKD and ASHD.

Physics-Informed Graph Neural Networks to solve 1-D equations of blood flow

BACKGROUND AND OBJECTIVE

Computational models of hemodynamics can contribute to optimizing surgical plans, and improve our understanding of cardiovascular diseases. Recently, machine learning methods have become essential to reduce the computational cost of these models. In this study, we propose a method that integrates 1-D blood flow equations with Physics-Informed Graph Neural Networks (PIGNNs) to estimate the propagation of blood flow velocity and lumen area pulse waves along arteries.

METHODS

Our methodology involves the creation of a graph based on arterial topology, where each 1-D line represents edges and nodes in the blood flow analysis. The innovation lies in decoding the mathematical data connecting the nodes, where each node has velocity and lumen area pulse waveform outputs. The training protocol for PIGNNs involves measurement data, specifically velocity waves measured from inlet and outlet vessels and diastolic lumen area measurements from each vessel. To optimize the learning process, our approach incorporates fundamental physical principles directly into the loss function. This comprehensive training strategy not only harnesses the power of machine learning but also ensures that PIGNNs respect fundamental laws governing fluid dynamics.

RESULTS

The accuracy was validated in silico with different arterial networks, where PIGNNs achieved a coefficient of determination (R2) consistently above 0.99, comparable to numerical methods like the discontinuous Galerkin scheme. Moreover, with in vivo data, the prediction reached R2 values greater than 0.80, demonstrating the method's effectiveness in predicting flow and lumen dynamics using minimal data.

CONCLUSIONS

This study showcased the ability to calculate lumen area and blood flow rate in blood vessels within a given topology by seamlessly integrating 1-D blood flow with PIGNNs, using only blood flow velocity measurements. Moreover, this study is the first to compare the PIGNNs method with other classic Physics-Informed Neural Network (PINNs) approaches for blood flow simulation. Our findings highlight the potential to use this cost-effective and proficient tool to estimate real-time arterial pulse waves.

Association between serum Klotho levels and estimated pulse wave velocity in postmenopausal women: a cross-sectional study of NHANES 2007-2016

Background

Postmenopausal women are at an increased risk of arterial stiffness, which can be assessed using estimated pulse wave velocity (ePWV). This study aimed to investigate the relationship between serum klotho levels and ePWV in postmenopausal women.

Methods

This cross-sectional study used data from postmenopausal women who participated in the National Health and Nutrition Examination Survey (NHANES) between 2007 and 2016. Participants were divided into two groups based on the presence of hypertension. Weighted multivariate linear regression was used to analyze the relationship between serum Klotho levels and ePWV in each group. Restricted cubic spline models with multivariable adjustments were employed to examine nonlinear associations within each group.

Results

Our analysis included 4,468 postmenopausal women from the NHANES database, with 1,671 in the non-hypertensive group and 2,797 in the hypertensive group. In all regression models, serum Klotho (ln-transformed) levels were significantly and independently negatively correlated with ePWV in the non-hypertensive group. After fully adjusting for confounders, a 1-unit increase in ln(Klotho) was associated with a 0.13 m/s decrease in ePWV (β = -0.13, 95% CI -0.23 to -0.03; p = 0.008). Additionally, in the fully adjusted model, participants in the highest quartile of ln(Klotho) had an ePWV value 0.14 m/s lower than those in the lowest quartile (p for trend = 0.017; 95% CI -0.23 to -0.05; p = 0.002). This negative correlation was consistent across subgroups and was particularly significant among women aged < 60 years, nonsmokers, and non-Hispanic Black women. However, no association was observed between serum Klotho levels and ePWV in the hypertensive group.

Conclusion

Hypertension may affect the relationship between serum Klotho level and ePWV in postmenopausal women. Increased serum Klotho levels may reduce arterial stiffness in postmenopausal women. Further studies are required to confirm these findings.

A new methodology for determining the central pressure waveform from peripheral measurement using Fourier-based machine learning

Radial applanation tonometry is a well-established technique for hemodynamic monitoring and is becoming popular in affordable non-invasive wearable healthcare electronics. To assess the central aortic pressure using radial-based measurements, there is an essential need to develop mathematical approaches to estimate the central pressure waveform. In this study, we propose a new Fourier-based machine learning (F-ML) methodology to transfer non-invasive radial pressure measurements to the central pressure waveform. To test the method, collection of tonometry recordings of the radial and carotid pressure measurements are used from the Framingham Heart Study (2640 individuals, 55 % women) with mean (range) age of 66 (40-91) years. Method-derived estimates are significantly correlated with the measured ones for three major features of the pressure waveform (systolic blood pressure, r=0.97, p < 0.001; diastolic blood pressure, r=0.99, p < 0.001; and mean blood pressure, r=0.99, p < 0.001). In all cases, the Bland-Altman analysis shows negligible bias in the estimations and error is bounded to 5.4 mmHg. Findings also suggest that the F-ML approach reconstructs the shape of the central pressure waveform accurately with the average normalized root mean square error of 5.5 % in the testing population which is blinded to all stages of machine learning development. The results show that the F-ML transfer function outperforms the conventional generalized transfer function, particularly in terms of reconstructing the shape of the central pressure waveform morphology. The proposed F-ML transfer function can provide accurate estimates for the central pressure waveform, and ultimately expand the usage of non-invasive devices for central hemodynamic assessment.

Estimation of aortic stiffness by finger photoplethysmography using enhanced pulse wave analysis and machine learning

Introduction

Aortic stiffness plays a critical role in the evolution of cardiovascular diseases, but the assessment requires specialized equipment. Photoplethysmography (PPG) and single-lead electrocardiogram (ECG) are readily available in healthcare and wearable devices. We studied whether a brief PPG registration, alone or in combination with single-lead ECG, could be used to reliably estimate aortic stiffness.

Methods

A proof-of-concept study with simultaneous high-resolution index finger recordings of infrared PPG, single-lead ECG, and finger blood pressure (Finapres) was performed in 33 participants [median age 44 (range 21-66) years, 19 men] and repeated within 2 weeks. Carotid-femoral pulse wave velocity (cfPWV; two-site tonometry with SphygmoCor) was used as a reference. A brachial single-cuff oscillometric device assessed aortic pulse wave velocity (aoPWV; Arteriograph) for further comparisons. We extracted 136 established PPG waveform features and engineered 13 new with improved coupling to the finger blood pressure curve. Height-normalized pulse arrival time (NPAT) was derived using ECG. Machine learning methods were used to develop prediction models.

Results

The best PPG-based models predicted cfPWV and aoPWV well (root-mean-square errors of 0.70 and 0.52 m/s, respectively), with minor improvements by adding NPAT. Repeatability and agreement were on par with the reference equipment. A new PPG feature, an amplitude ratio from the early phase of the waveform, was most important in modelling, showing strong correlations with cfPWV and aoPWV (r = -0.81 and -0.75, respectively, both P < 0.001).

Conclusion

Using new features and machine learning methods, a brief finger PPG registration can estimate aortic stiffness without requiring additional information on age, anthropometry, or blood pressure. Repeatability and agreement were comparable to those obtained using non-invasive reference equipment. Provided further validation, this readily available simple method could improve cardiovascular risk evaluation, treatment, and prognosis.

Update on the Use of Pulse Wave Velocity to Measure Age-Related Vascular Changes

PURPOSE OF REVIEW

Pulse wave velocity (PWV) is an important and well-established measure of arterial stiffness that is strongly associated with aging. Age-related alterations in the elastic properties and integrity of arterial walls can lead to cardiovascular disease. PWV measurements play an important role in the early detection of these changes, as well as other cardiovascular disease risk factors, such as hypertension. This review provides a comprehensive summary of the current knowledge of the effects of aging on arterial stiffness, as measured by PWV.

RECENT FINDINGS

This review highlights recent findings showing the applicability of PWV analysis for investigating heart failure, hypertension, and other cardiovascular diseases, as well as cerebrovascular diseases and Alzheimer's disease. It also discusses the clinical implications of utilizing PWV to monitor treatment outcomes, various challenges in implementing PWV assessment in clinical practice, and the development of new technologies, including machine learning and artificial intelligence, which may improve the usefulness of PWV measurements in the future. Measuring arterial stiffness through PWV remains an important technique to study aging, especially as the technology continues to evolve. There is a clear need to leverage PWV to identify interventions that mitigate age-related increases in PWV, potentially improving CVD outcomes and promoting healthy vascular aging.

Arterial Stiffness: From Basic Primers to Integrative Physiology

The elastic properties of conductance arteries are one of the most important hemodynamic functions in the body, and data continue to emerge regarding the importance of their dysfunction in vascular aging and a range of cardiovascular diseases. Here, we provide new insight into the integrative physiology of arterial stiffening and its clinical consequence. We also comprehensively review progress made on pathways/molecules that appear today as important basic determinants of arterial stiffness, particularly those mediating the vascular smooth muscle cell (VSMC) contractility, plasticity and stiffness. We focus on membrane and nuclear mechanotransduction, clearance function of the vascular wall, phenotypic switching of VSMCs, immunoinflammatory stimuli and epigenetic mechanisms. Finally, we discuss the most important advances of the latest clinical studies that revisit the classical therapeutic concepts of arterial stiffness and lead to a patient-by-patient strategy according to cardiovascular risk exposure and underlying disease.

2024 Recommendations for Validation of Noninvasive Arterial Pulse Wave Velocity Measurement Devices

BACKGROUND

Arterial stiffness, as measured by arterial pulse wave velocity (PWV), is an established biomarker for cardiovascular risk and target-organ damage in individuals with hypertension. With the emergence of new devices for assessing PWV, it has become evident that some of these devices yield results that display significant discrepancies compared with previous devices. This discrepancy underscores the importance of comprehensive validation procedures and the need for international recommendations.

METHODS

A stepwise approach utilizing the modified Delphi technique, with the involvement of key scientific societies dedicated to arterial stiffness research worldwide, was adopted to formulate, through a multidisciplinary vision, a shared approach to the validation of noninvasive arterial PWV measurement devices.

RESULTS

A set of recommendations has been developed, which aim to provide guidance to clinicians, researchers, and device manufacturers regarding the validation of new PWV measurement devices. The intention behind these recommendations is to ensure that the validation process can be conducted in a rigorous and consistent manner and to promote standardization and harmonization among PWV devices, thereby facilitating their widespread adoption in clinical practice.

CONCLUSIONS

It is hoped that these recommendations will encourage both users and developers of PWV measurement devices to critically evaluate and validate their technologies, ultimately leading to improved consistency and comparability of results. This, in turn, will enhance the clinical utility of PWV as a valuable tool for assessing arterial stiffness and informing cardiovascular risk stratification and management in individuals with hypertension.

Noninvasive hemodynamic indices of vascular aging: an in silico assessment

Vascular aging (VA) involves structural and functional changes in blood vessels that contribute to cardiovascular disease. Several noninvasive pulse wave (PW) indices have been proposed to assess the arterial stiffness component of VA in the clinic and daily life. This study investigated 19 of these indices, identified in recent review articles on VA, by using a database comprising 3,837 virtual healthy subjects aged 25-75 yr, each with unique PW signals simulated under various levels of artificial noise to mimic real measurement errors. For each subject, VA indices were calculated from filtered PW signals and compared with the precise theoretical value of aortic Young's modulus (EAo). In silico PW indices showed age-related changes that align with in vivo population studies. The cardio-ankle vascular index (CAVI) and all pulse wave velocity (PWV) indices showed strong linear correlations with EAo (Pearson's rp > 0.95). Carotid distensibility showed a strong negative nonlinear correlation (Spearman's rs < -0.99). CAVI and distensibility exhibited greater resilience to noise compared with PWV indices. Blood pressure-related indices and photoplethysmography (PPG)-based indices showed weaker correlations with EAo (rp and rs < 0.89, |rp| and |rs| < 0.84, respectively). Overall, blood pressure-related indices were confounded by more cardiovascular properties (heart rate, stroke volume, duration of systole, large artery diameter, and/or peripheral vascular resistance) compared with other studied indices, and PPG-based indices were most affected by noise. In conclusion, carotid-femoral PWV, CAVI and carotid distensibility emerged as the superior clinical VA indicators, with a strong EAo correlation and noise resilience. PPG-based indices showed potential for daily VA monitoring under minimized noise disturbances.NEW & NOTEWORTHY For the first time, 19 noninvasive pulse wave indices for assessing vascular aging were examined together in a single database of nearly 4,000 subjects aged 25-75 yr. The dataset contained precise values of the aortic Young's modulus and other hemodynamic measures for each subject, which enabled us to test each index's ability to measure changes in aortic stiffness while accounting for confounding factors and measurement errors. The study provides freely available tools for analyzing these and additional indices.

Use of a rat model to characterize 35 arterial pulse wave parameters in a comparative study of isoflurane and Zoletil/xylazine anesthesia and the effect of Acanthopanax senticosus extract

BACKGROUND

Information obtained from arterial pulse waveforms (APW) can be useful for characterizing the cardiovascular system. To achieve this, it is necessary to know the detailed characteristics of APWs in different states of an organism, which would allow APW parameters (APW-Ps) to be assigned to particular (patho)physiological conditions. Therefore, our work aimed to characterize 35 APW-Ps in rats under the influence of isoflurane (ISO) and Zoletil/xylazine (ZO/XY) anesthesia and to study the effect of root extract from Acanthopanax senticosus (ASRE) in these anesthetic conditions.

METHODS

The right jugular vein of anesthetized rats was cannulated for the administration of ASRE and the left carotid artery for the detection of APWs from which 35 APW-Ps were evaluated.

RESULTS

We obtained data on 35 APW-Ps, which significantly depended on the anesthesia, and thus, they characterized the cardiovascular system under these two conditions. ASRE transiently modulated all 35 APW-Ps, including a transient decrease in systolic and diastolic blood pressure (BP) and heart rate or increases in pulse BP, dP/dtmax , and systolic and diastolic areas. Whereas the transient effects of ASRE were similar, the extract had prolonged disturbing effects on the cardiovascular system in rats under ZO/XY but not under ISO anesthesia. This negative effect can result from the disturbance caused by ZO/XY anesthesia on the cardiovascular system.

CONCLUSIONS

We characterized 35 APW-Ps of rats under ISO and ZO/XY anesthesia and found that ASRE contains compounds that can modulate the properties of the cardiovascular system, which significantly depended on the status of the cardiovascular system. This should be considered when using ASRE as a nutritional supplement by individuals with cardiovascular problems.

The Phthalic Selenoanhydride Decreases Rat Blood Pressure and Tension of Isolated Mesenteric, Femoral and Renal Arteries

Phthalic selenoanhydride (R-Se) solved in physiological buffer releases various reactive selenium species including H₂Se. It is a potential compound for Se supplementation which exerts several biological effects, but its effect on the cardiovascular system is still unknown. Therefore, herein we aimed to study how R-Se affects rat hemodynamic parameters and vasoactive properties in isolated arteries. The right jugular vein of anesthetized Wistar male rats was cannulated for IV administration of R-Se. The arterial pulse waveform (APW) was detected by cannulation of the left carotid artery, enabling the evaluation of 35 parameters. R-Se (1-2 µmol kg^-1), but not phthalic anhydride or phthalic thioanhydride, transiently modulated most of the APW parameters including a decrease in systolic and diastolic blood pressure, heart rate, dP/dt_max relative level, or anacrotic/dicrotic notches, whereas systolic area, dP/dt_min delay, dP/dt_d delay, anacrotic notch relative level or its delay increased. R-Se (~10-100 µmol L^-1) significantly decreased the tension of precontracted mesenteric, femoral, and renal arteries, whereas it showed a moderate vasorelaxation effect on thoracic aorta isolated from normotensive Wistar rats. The results imply that R-Se acts on vascular smooth muscle cells, which might underlie the effects of R-Se on the rat hemodynamic parameters.

Central Arterial Dynamic Evaluation from Peripheral Blood Pressure Waveforms Using CycleGAN: An In Silico Approach

Arterial stiffness is a major condition related to many cardiovascular diseases. Traditional approaches in the assessment of arterial stiffness supported by machine learning techniques are limited to the pulse wave velocity (PWV) estimation based on pressure signals from the peripheral arteries. Nevertheless, arterial stiffness can be assessed based on the pressure-strain relationship by analyzing its hysteresis loop. In this work, the capacity of deep learning models based on generative adversarial networks (GANs) to transfer pressure signals from the peripheral arterial region to pressure and area signals located in the central arterial region is explored. The studied signals are from a public and validated virtual database. Compared to other works in which the assessment of arterial stiffness was performed via PWV, in the present work the pressure-strain hysteresis loop is reconstructed and evaluated in terms of classical machine learning metrics and clinical parameters. Least-square GAN (LSGAN) and Wasserstein GAN with gradient penalty (WGAN-GP) adversarial losses are compared, yielding better results with LSGAN. LSGAN mean ± standard deviation of error for pressure and area pulse waveforms are 0.8 ± 0.4 mmHg and 0.1 ± 0.1 cm², respectively. Regarding the pressure-strain elastic modulus, it is achieved a mean absolute percentage error of 6.5 ± 5.1%. GAN-based deep learning models can recover the pressure-strain loop of central arteries while observing pressure signals from peripheral arteries.

A learning-based image processing approach for pulse wave velocity estimation using spectrogram from peripheral pulse wave signals: An in silico study

Carotid-to-femoral pulse wave velocity (cf-PWV) is considered a critical index to evaluate arterial stiffness. For this reason, estimating Carotid-to-femoral pulse wave velocity (cf-PWV) is essential for diagnosing and analyzing different cardiovascular diseases. Despite its broader adoption in the clinical routine, the measurement process of carotid-to-femoral pulse wave velocity is considered a demanding task for clinicians and patients making it prone to inaccuracies and errors in the estimation. A smart non-invasive, and peripheral measurement of carotid-to-femoral pulse wave velocity could overcome the challenges of the classical assessment process and improve the quality of patient care. This paper proposes a novel methodology for the carotid-to-femoral pulse wave velocity estimation based on the use of the spectrogram representation from single non-invasive peripheral pulse wave signals [photoplethysmography (PPG) or blood pressure (BP)]. This methodology was tested using three feature extraction methods based on the semi-classical signal analysis (SCSA) method, the Law's mask for texture energy extraction, and the central statistical moments. Finally, each feature method was fed into different machine learning models for the carotid-to-femoral pulse wave velocity estimation. The proposed methodology obtained an $R²\geq0.90$ for all the peripheral signals for the noise-free case using the MLP model, and for the different noise levels added to the original signal, the SCSA-based features with the MLP model presented an $R²\geq0.91$ for all the peripheral signals at the level of noise. These results provide evidence of the capacity of spectrogram representation for efficiently assessing the carotid-to-femoral pulse wave velocity estimation using different feature methods. Future work will be done toward testing the proposed methodology for in-vivo signals.

Machine Learning for Aiding Blood Flow Velocity Estimation Based on Angiography

Computational fluid dynamics (CFD) is widely employed to predict hemodynamic characteristics in arterial models, while not friendly to clinical applications due to the complexity of numerical simulations. Alternatively, this work proposed a framework to estimate hemodynamics in vessels based on angiography images using machine learning (ML) algorithms. First, the iodine contrast perfusion in blood was mimicked by a flow of dye diffusing into water in the experimentally validated CFD modeling. The generated projective images from simulations imitated the counterpart of light passing through the flow field as an analogy of X-ray imaging. Thus, the CFD simulation provides both the ground truth velocity field and projective images of dye flow patterns. The rough velocity field was estimated using the optical flow method (OFM) based on 53 projective images. ML training with least absolute shrinkage, selection operator and convolutional neural network was conducted with CFD velocity data as the ground truth and OFM velocity estimation as the input. The performance of each model was evaluated based on mean absolute error and mean squared error, where all models achieved or surpassed the criteria of 3 × 10-3 and 5 × 10-7 m/s, respectively, with a standard deviation less than 1 × 10-6 m/s. Finally, the interpretable regression and ML models were validated with over 613 image sets. The validation results showed that the employed ML model significantly reduced the error rate from 53.5% to 2.5% on average for the v-velocity estimation in comparison with CFD. The ML framework provided an alternative pathway to support clinical diagnosis by predicting hemodynamic information with high efficiency and accuracy.

A Multiple Linear Regression Model for Carotid-to-Femoral Pulse Wave Velocity Estimation Based on Schrodinger Spectrum Characterization

In this paper, a multiple linear regression model for estimating the Carotid-to-femoral pulse wave velocity (cf-PWV) from a single non-invasive peripheral pulse wave, namely blood pressure or photoplethysmography, is proposed. The training and testing datasets were extracted from in-silico, publicly available, pulse waves and hemodynamics data. The proposed model relies on a preprocessing and features extraction steps, which are performed using a semi-classical signal analysis (SCSA) method. The obtained results provide more evidence for the feasibility of machine learning and the SCSA method as a smart tool for the efficient assessment of the cf-PWV.

Wave Intensity Analysis Combined With Machine Learning can Detect Impaired Stroke Volume in Simulations of Heart Failure

Heart failure is treatable, but in the United Kingdom, the 1-, 5- and 10-year mortality rates are 24.1, 54.5 and 75.5%, respectively. The poor prognosis reflects, in part, the lack of specific, simple and affordable diagnostic techniques; the disease is often advanced by the time a diagnosis is made. Previous studies have demonstrated that certain metrics derived from pressure-velocity-based wave intensity analysis are significantly altered in the presence of impaired heart performance when averaged over groups, but to date, no study has examined the diagnostic potential of wave intensity on an individual basis, and, additionally, the pressure waveform can only be obtained accurately using invasive methods, which has inhibited clinical adoption. Here, we investigate whether a new form of wave intensity based on noninvasive measurements of arterial diameter and velocity can detect impaired heart performance in an individual. To do so, we have generated a virtual population of two-thousand elderly subjects, modelling half as healthy controls and half with an impaired stroke volume. All metrics derived from the diameter-velocity-based wave intensity waveforms in the carotid, brachial and radial arteries showed significant crossover between groups-no one metric in any artery could reliably indicate whether a subject's stroke volume was normal or impaired. However, after applying machine learning to the metrics, we found that a support vector classifier could simultaneously achieve up to 99% recall and 95% precision. We conclude that noninvasive wave intensity analysis has significant potential to improve heart failure screening and diagnosis.

A proof of concept study for machine learning application to stenosis detection

This proof of concept (PoC) assesses the ability of machine learning (ML) classifiers to predict the presence of a stenosis in a three vessel arterial system consisting of the abdominal aorta bifurcating into the two common iliacs. A virtual patient database (VPD) is created using one-dimensional pulse wave propagation model of haemodynamics. Four different machine learning (ML) methods are used to train and test a series of classifiers—both binary and multiclass—to distinguish between healthy and unhealthy virtual patients (VPs) using different combinations of pressure and flow-rate measurements. It is found that the ML classifiers achieve specificities larger than 80% and sensitivities ranging from 50 to 75%. The most balanced classifier also achieves an area under the receiver operative characteristic curve of 0.75, outperforming approximately 20 methods used in clinical practice, and thus placing the method as moderately accurate. Other important observations from this study are that (i) few measurements can provide similar classification accuracies compared to the case when more/all the measurements are used; (ii) some measurements are more informative than others for classification; and (iii) a modification of standard methods can result in detection of not only the presence of stenosis, but also the stenosed vessel.

Graphical Abstract