Selective Partitioned Regression for Accurate Kidney Health Monitoring.

The number of people diagnosed with advanced stages of kidney disease have been rising every year. Early detection and constant monitoring are the only minimally invasive means to prevent severe kidney damage or kidney failure. We propose a cost-effective machine learning-based testing system that can facilitate inexpensive yet accurate kidney health checks. Our proposed framework, which was developed into an iPhone application, uses a camera-based bio-sensor and state-of-the-art classical machine learning and deep learning techniques for predicting the concentration of creatinine in the sample, based on colorimetric change in the test strip. The predicted creatinine concentration is then used to classify the severity of the kidney disease as healthy, intermediate, or critical. In this article, we focus on the effectiveness of machine learning models to translate the colorimetric reaction to kidney health prediction. In this setting, we thoroughly evaluated the effectiveness of our novel proposed models against state-of-the-art classical machine learning and deep learning approaches. Additionally, we executed a number of ablation studies to measure the performance of our model when trained using different meta-parameter choices. Our evaluation results indicate that our selective partitioned regression (SPR) model, using histogram of colors-based features and a histogram gradient boosted trees underlying estimator, exhibits much better overall prediction performance compared to state-of-the-art methods. Our initial study indicates that SPR can be an effective tool for detecting the severity of kidney disease using inexpensive lateral flow assay test strips and a smart phone-based application. Additional work is needed to verify the performance of the model in various settings.

Estrogen receptor-α prevents right ventricular diastolic dysfunction and fibrosis in female rats.

Although women are more susceptible to pulmonary arterial hypertension (PAH) than men, their right ventricular (RV) function is better preserved. Estrogen receptor-α (ERα) has been identified as a likely mediator for estrogen protection in the RV. However, the role of ERα in preserving RV function and remodeling during pressure overload remains poorly understood. We hypothesized that loss of functional ERα removes female protection from adverse remodeling and is permissive for the development of a maladapted RV phenotype. Male and female rats with a loss-of-function mutation in ERα (ERαMut) and wild-type (WT) littermates underwent RV pressure overload by pulmonary artery banding (PAB). At 10 wk post-PAB, WT and ERαMut demonstrated RV hypertrophy. Analysis of RV pressure waveforms demonstrated RV-pulmonary vascular uncoupling and diastolic dysfunction in female, but not male, ERαMut PAB rats. Similarly, female, but not male, ERαMut exhibited increased RV fibrosis, comprised primarily of thick collagen fibers. There was an increased protein expression ratio of TIMP metallopeptidase inhibitor 1 (Timp1) to matrix metalloproteinase 9 (Mmp9) in female ERαMut compared with WT PAB rats, suggesting less collagen degradation. RNA-sequencing in female WT and ERαMut RV revealed kallikrein-related peptidase 10 (Klk10) and Jun Proto-Oncogene (Jun) as possible mediators of female RV protection during PAB. In summary, ERα in females is protective against RV-pulmonary vascular uncoupling, diastolic dysfunction, and fibrosis in response to pressure overload. ERα appears to be dispensable for RV adaptation in males. ERα may be a mediator of superior RV adaptation in female patients with PAH.NEW & NOTEWORTHY Using a novel loss-of-function mutation in estrogen receptor-α (ERα), we demonstrate that female, but not male, ERα mutant rats display right ventricular (RV)-vascular uncoupling, diastolic dysfunction, and fibrosis following pressure overload, indicating a sex-dependent role of ERα in protecting against adverse RV remodeling. TIMP metallopeptidase inhibitor 1 (Timp1), matrix metalloproteinase 9 (Mmp9), kallikrein-related peptidase 10 (Klk10), and Jun Proto-Oncogene (Jun) were identified as potential mediators in ERα-regulated pathways in RV pressure overload.

A novel single-beat approach to assess right ventricular systolic function.

Clinical assessment of right ventricular (RV) contractility in diseases such as pulmonary arterial hypertension (PAH) has been hindered by the lack of a robust methodology. Here, a novel, clinically viable, single-beat method was developed to assess end-systolic elastance (Ees), a measure of right ventricular (RV) contractility. We hypothesized that this novel approach reduces uncertainty and interobserver variability in the estimation of the maximum isovolumic pressure (Piso), the key step in single-beat methods. The new method was designed to include a larger portion of the RV pressure data and minimize subjective adjustments by the operator. Data were obtained from right heart catheterization of PAH patients in a multicenter prospective study ( data set 1) and a single-center retrospective study ( data set 2). To obtain Piso, three independent observers used an established single-beat method (based on the first derivative of the pressure waveform) and the novel method (based on the second derivative). Interobserver variability analysis included paired t-test, one-way ANOVA, interclass correlation (ICC) analysis, and a modified Bland-Altman analysis. The Piso values obtained from the two methods were linearly correlated for both data set 1 ( R2 = 0.74) and data set 2 ( R2 = 0.91). Compared with the established method, the novel method resulted in smaller interobserver variability ( P < 0.001), nonsignificant differences between observers, and a narrower confidence interval. By reducing uncertainty and interobserved variability, this novel approach may pave the way for more effective clinical management of PAH. NEW & NOTEWORTHY A novel methodology to assess right ventricular contractility from clinical data is demonstrated. This approach significantly reduces interobserver variability in the analysis of ventricular pressure data, as demonstrated in a relatively large population of subjects with pulmonary hypertension. This study may enable more accurate clinical monitoring of systolic function in subjects with pulmonary hypertension.

Dobutamine stress MRI in pulmonary hypertension: relationships between stress pulmonary artery relative area change, RV performance, and 10-year survival.

In pulmonary hypertension (PH), right ventricular (RV) performance determines survival. Pulmonary artery (PA) stiffening is an important biomechanical event in PH and also predicts survival based on the PA relative area change (RAC) measured at rest using magnetic resonance imaging (MRI). In this exploratory study, we sought to generate novel hypotheses regarding the influence of stress RAC on PH prognosis and the interaction between PA stiffening, RV performance and survival. Fifteen PH patients underwent dobutamine stress-MRI (ds-MRI) and right heart catheterization. RACREST, RACSTRESS, and ΔRAC (RAC STRESS - RAC REST) were correlated against resting invasive hemodynamics and ds-MRI data regarding RV performance and RV-PA coupling efficiency (n'vv [RV stroke volume/RV end-systolic volume]). The impact of RAC, RV data, and n'vv on ten-year survival were determined using Kaplan-Meier analysis. PH patients with a low ΔRAC (<-2.6%) had a worse long-term survival (log-rank P = 0.045, HR for death = 4.46 [95% CI = 1.08-24.5]) than those with ΔRAC ≥ -2.6%. Given the small sample, these data should be interpreted with caution; however, low ΔRAC was associated with an increase in stress diastolic PA area indicating proximal PA stiffening. Associations of borderline significance were observed between low RACSTRESS and low n'vvSTRESS, Δη'VV, and ΔRVEF. Further studies are required to validate the potential prognostic impact of ΔRAC and the biomechanics potentially connecting low ΔRAC to shorter survival. Such studies may facilitate development of novel PH therapies targeted to the proximal PA.

Reduced haemodynamic coupling and exercise are associated with vascular stiffening in pulmonary arterial hypertension.

OBJECTIVE: Inadequate right ventricular (RV) and pulmonary arterial (PA) functional responses to exercise are important yet poorly understood features of pulmonary arterial hypertension (PAH). This study combined invasive catheterisation with echocardiography to assess RV afterload, RV function and ventricular-vascular coupling in subjects with PAH. METHODS: Twenty-six subjects with PAH were prospectively recruited to undergo right heart catheterisation and Doppler echocardiography at rest and during incremental exercise, and cardiac MRI at rest. Measurements at rest included basic haemodynamics, RV function and coupling efficiency (η). Measurements during incremental exercise included pulmonary vascular resistance (Z0), characteristic impedance (ZC, a measure of proximal PA stiffness) and proximal and distal PA compliance (CPA). RESULTS: In patients with PAH, the proximal PAs were significantly stiffer at maximum exercise (ZC =2.31±0.38 vs 1.33±0.15 WU×m2 at rest; p=0.003) and PA compliance was decreased (CPA=0.88±0.10 vs 1.32±0.17 mL/mm Hg/m2 at rest; p=0.0002). Z0 did not change with exercise. As a result, the resistance-compliance (RC) time decreased with exercise (0.67±0.05 vs 1.00±0.07 s at rest; p<10-6). When patients were grouped according to resting coupling efficiency, those with poorer η exhibited stiffer proximal PAs at rest, a lower maximum exercise level, and more limited CPA reduction at maximum exercise. CONCLUSIONS: In PAH, exercise causes proximal and distal PA stiffening, which combined with preserved Z0 results in decreased RC time with exercise. Stiff PAs at rest may also contribute to poor haemodynamic coupling, reflecting reduced pulmonary vascular reserve that contributes to limit the maximum exercise level tolerated.

Imaging right ventricular function to predict outcome in pulmonary arterial hypertension.

BACKGROUND: Right ventricular (RV) function is a major determinant of outcome in pulmonary arterial hypertension (PAH). However, uncertainty persists about the optimal method of evaluation. METHODS: We measured RV end-systolic and end-diastolic volumes (ESV and EDV) using cardiac magnetic resonance imaging and RV pressures during right heart catheterization in 140 incident PAH patients and 22 controls. A maximum RV pressure (Pmax) was calculated from the nonlinear extrapolations of early and late systolic portions of the RV pressure curve. The gold standard measure of RV function adaptation to afterload, or RV-arterial coupling (Ees/Ea) was estimated by the stroke volume (SV)/ESV ratio (volume method) or as Pmax/mean pulmonary artery pressure (mPAP) minus 1 (pressure method) (n=84). RV function was also assessed by ejection fraction (EF), right atrial pressure (RAP) and SV. RESULTS: Higher Ea and RAP, and lower compliance, SV and EF predicted outcome at univariate analysis. Ees/Ea estimated by the pressure method did not predict outcome but Ees/Ea estimated by the volume method (SV/ESV) did. At multivariate analysis, only SV/ESV and EF were independent predictors of outcome. Survival was poorer in patients with a fall in EF or SV/ESV during follow-up (n=44, p=0.008). CONCLUSION: RV function to predict outcome in PAH is best evaluated by imaging derived SV/ESV or EF. In this study, there was no added value of invasive measurements or simplified pressure-derived estimates of RV-arterial coupling.

High-Resolution Measurements of Velocity and Shear Stress in Leakage Jets From Bileaflet Mechanical Heart Valve Hinge Models.

In flow through cardiovascular implants, hemolysis, and thrombosis may be initiated by nonphysiological shear stress on blood elements. To enhance understanding of the small-scale flow structures that stimulate cellular responses, and ultimately to design devices for reduced blood damage, it is necessary to study the flow-field at high spatial and temporal resolution. In this work, we investigate flow in the reverse leakage jet from the hinge of a bileaflet mechanical heart valve (BMHV). Scaled-up model hinges are employed, enabling measurement of the flow-field at effective spatial resolution of 167 µm and temporal resolution of 594 µs using two-component particle image velocimetry (PIV). High-velocity jets were observed at the hinge outflow, with time-average velocity up to 5.7 m/s, higher than reported in previous literature. Mean viscous shear stress is up to 60 Pa. For the first time, strongly unsteady flow has been observed in the leakage jet. Peak instantaneous shear stress is up to 120 Pa, twice as high as the average value. These high-resolution measurements identify the hinge leakage jet as a region of very high fluctuating shear stress which is likely to be thrombogenic and should be an important target for future design improvement.

Accuracy of Doppler echocardiographic estimates of pulmonary artery pressures in a canine model of pulmonary hypertension.

OBJECTIVES: To compare noninvasive estimates of pulmonary artery pressure (PAP) obtained via echocardiography (ECHO) to invasive measurements of PAP obtained during right heart catheterization (RHC) across a range of PAP. To examine the accuracy of estimating right atrial pressure via ECHO (RAPECHO) compared to RAP measured by RHC (RAPRHC), and determine if adding RAPECHO improves the accuracy of noninvasive PAP estimations. ANIMALS: 14 healthy female beagle dogs. METHODS: Comparison of ECHO and RHC measures of PAP, both at normal PAP and increased PAP generated by microbead embolization. RESULTS: Noninvasive estimates of PAP were moderately but significantly correlated (r of 0.68-0.78; p < 0.0006) with invasive measurements of PAP. Wide variance was noted for all estimations, with increased variance at higher PAP. The addition of RAPECHO improved correlation and bias in all cases. RAPRHC was significantly correlated with RAPECHO (r = 0.38; p = 0.04) as estimated by the ellipse area method. Median RAPRHC was significantly different between 3 subjective assessments of right atrial size (p = 0.037). CONCLUSIONS: Spectral Doppler assessments of tricuspid and pulmonic regurgitation are imperfect methods for predicting PAP as measured by catheterization despite an overall moderate correlation between invasive and noninvasive values. Noninvasive measurements may be better utilized as part of a comprehensive assessment of PAP in canine patients. RAPRHC appears best estimated based on subjective assessment of RA size. Including estimated RAPECHO in estimates of PAP improves the correlation and relatedness between noninvasive and invasive measures of PAP, but notable variability in accuracy of estimations persists.

Pulmonary artery relative area change is inversely related to ex vivo measured arterial elastic modulus in the canine model of acute pulmonary embolization.

A low relative area change (RAC) of the proximal pulmonary artery (PA) over the cardiac cycle is a good predictor of mortality from right ventricular failure in patients with pulmonary hypertension (PH). The relationship between RAC and local mechanical properties of arteries, which are known to stiffen in acute and chronic PH, is not clear, however. In this study, we estimated elastic moduli of three PAs (MPA, LPA and RPA: main, left and right PAs) at the physiological state using mechanical testing data and correlated these estimated elastic moduli to RAC measured in vivo with both phase-contrast magnetic resonance imaging (PC-MRI) and M-mode echocardiography (on RPA only). We did so using data from a canine model of acute PH due to embolization to assess the sensitivity of RAC to changes in elastic modulus in the absence of chronic PH-induced arterial remodeling. We found that elastic modulus increased with embolization-induced PH, presumably a consequence of increased collagen engagement, which corresponds well to decreased RAC. Furthermore, RAC was inversely related to elastic modulus. Finally, we found MRI and echocardiography yielded comparable estimates of RAC. We conclude that RAC of proximal PAs can be obtained from either MRI or echocardiography and a change in RAC indicates a change in elastic modulus of proximal PAs detectable even in the absence of chronic PH-induced arterial remodeling. The correlation between RAC and elastic modulus of proximal PAs may be useful for prognoses and to monitor the effects of therapeutic interventions in patients with PH.

Methods for measuring right ventricular function and hemodynamic coupling with the pulmonary vasculature.

The right ventricle (RV) is a pulsatile pump, the efficiency of which depends on proper hemodynamic coupling with the compliant pulmonary circulation. The RV and pulmonary circulation exhibit structural and functional differences with the more extensively investigated left ventricle (LV) and systemic circulation. In light of these differences, metrics of LV function and efficiency of coupling to the systemic circulation cannot be used without modification to characterize RV function and efficiency of coupling to the pulmonary circulation. In this article, we review RV physiology and mechanics, established and novel methods for measuring RV function and hemodynamic coupling, and findings from application of these methods to RV function and coupling changes with pulmonary hypertension. We especially focus on non-invasive measurements, as these may represent the future for clinical monitoring of disease progression and the effect of drug therapies.

Impact of acute pulmonary embolization on arterial stiffening and right ventricular function in dogs.

Pulmonary hypertension (PH) can impact right ventricular (RV) function and alter pulmonary artery (PA) stiffness. The response of the RV to an acute increase in pulmonary pressure is unclear. In addition, the relation between total pulmonary arterial compliance and local PA stiffness has not been investigated. We used a combination of right heart catheterization (RHC) and magnetic resonance imaging (MRI) to assess PA stiffening and RV function in dogs before and after acute embolization. We hypothesized that in moderate, acute PH the RV is able to compensate for increased afterload, maintaining adequate coupling. Also, we hypothesized that in the absence of PA remodeling the relative area change in the proximal PA (RAC, a noninvasive index of local area strain) correlates with the total arterial compliance (stroke volume-to-pulse pressure ratio). Our results indicate that, after embolization, RV function is able to accommodate the demand for increased stroke work without uncoupling, albeit at the expense of a reduction of efficiency. In this acute model, RAC showed excellent correlation with total arterial compliance. We used this correlation to assess PA pulse pressure (PP) from noninvasive MRI measurements of stroke volume and RAC. We demonstrated that in acute pulmonary embolism MRI estimates of PP are remarkably close to measurements from RHC. These results, if confirmed in chronic PH and clinically, suggest that monitoring of PH progression by noninvasive methods may be possible.

High-resolution measurement of the unsteady velocity field to evaluate blood damage induced by a mechanical heart valve.

We investigate the potential of prosthetic heart valves to generate abnormal flow and stress patterns, which can contribute to platelet activation and lysis according to blood damage accumulation mechanisms. High-resolution velocity measurements of the unsteady flow field, obtained with a standard particle image velocimetry system and a scaled-up model valve, are used to estimate the shear stresses arising downstream of the valve, accounting for flow features at scales less than one order of magnitude larger than blood cells. Velocity data at effective spatial and temporal resolution of 60 µm and 1.75 kHz, respectively, enabled accurate extraction of Lagrangian trajectories and loading histories experienced by blood cells. Non-physiological stresses up to 10 Pa were detected, while the development of vortex flow in the wake of the valve was observed to significantly increase the exposure time, favouring platelet activation. The loading histories, combined with empirical models for blood damage, reveal that platelet activation and lysis are promoted at different stages of the heart cycle. Shear stress and blood damage estimates are shown to be sensitive to measurement resolution.