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.

The Use of Maximum Entropy to Enhance Wave Intensity Analysis: An Application to Coronary Arteries in Hypertrophic Obstructive Cardiomyopathy

Background: Wave intensity analysis is useful for analyzing coronary hemodynamics. Much of its clinical application involves the identification of waves indicated by peaks in the wave intensity and relating their presence or absence to different cardiovascular events. However, the analysis of wave intensity peaks can be problematic because of the associated noise in the measurements. This study shows how wave intensity analysis can be enhanced by using a Maximum Entropy Method (MEM).Methods: We introduce a MEM to differentiate between “peaks” and “background” in wave intensity waveforms. We apply the method to the wave intensity waveforms measured in the left anterior descending coronary artery from 10 Hypertrophic Obstructive Cardiomyopathy (HOCM) and 11 Controls with normal cardiac function. We propose a naming convention for the significant waves and compare them across the cohorts.Results: Using a MEM enhances wave intensity analysis by identifying twice as many significant waves as previous studies. The results are robust when MEM is applied to the log transformed wave intensity data and when all of the measured data are used. Comparing waves across cohorts, we suggest that the absence of a forward expansion wave in HOCM can be taken as an indication of HOCM. Our results also indicate that the backward compression waves in HOCM are significantly larger than in Controls; unlike the forward compression waves where the wave energy in Controls is significantly higher than in HOCM. Comparing the smaller secondary waves revealed by MEM, we find some waves that are present in the majority of Controls and absent in almost all HOCM, and other waves that are present in some HOCM patients but entirely absent in Controls. This suggests some diagnostic utility in the clinical measurement of these waves, which can be a positive sign of HOCM or a subgroup with a particular pathology.Conclusion: The MEM enhances wave intensity analysis by identifying many more significant waves. The method is novel and can be applied to wave intensity analysis in all arteries. As an example, we show how it can be useful in the clinical study of hemodynamics in the coronary arteries in HOCM.

Feasibility of Wave Intensity Analysis in Patients With Conotruncal Anomalies Before and After Pregnancy: New Physiological Insights?

Background: Conotruncal anomalies (CTA) are associated with ongoing dilation of the aortic root, as well as increased aortic stiffness, which may relate to intrinsic properties of the aorta. Pregnancy hormones lead to hemodynamic changes and remodeling of the tunica media, resulting in the opposite effect, i.e., increasing distensibility. These changes normalize post-pregnancy in healthy women but have not been fully investigated in CTA patients.Methods: We examined aortic distensibility and ventriculo-arterial coupling before and after pregnancy using cardiovascular magnetic resonance (CMR)-derived wave intensity analysis (WIA). Pre- and post-pregnancy CMR data were retrospectively analyzed. Aortic diameters were measured before, during, and after pregnancy by cardiac ultrasound and before and after pregnancy by CMR. Phase contrast MR flow sequences were used for calculating wave speed (c) and intensity (WI). A matched analysis was performed comparing results before and after pregnancy.Results: Thirteen women (n = 5, transposition of the great arteries; n = 6, tetralogy of Fallot; n = 1, double outlet right ventricle, n = 1, truncus arteriosus) had 19 pregnancies. Median time between delivery and second CMR was 2.3 years (range: 1–6 years). The aortic diameter increased significantly after pregnancy in nine (n = 9) patients by a median of 4 ± 2.3 mm (range: 2–7.0 mm, p = 0.01). There was no difference in c pre-/post-pregnancy (p = 0.73), suggesting that increased compliance, typically observed during pregnancy, does not persist long term. A significant inverse relationship was observed between c and heart rate (HR) after pregnancy (p = 0.01, r = 0.73). There was no significant difference in cardiac output, aortic/pulmonary regurgitation, or WI peaks pre-/post-pregnancy.Conclusions: WIA is feasible in this population and could provide physiological insights in larger cohorts. Aortic distensibility and wave intensity did not change before and after pregnancy in CTA patients, despite an increase in diameter, suggesting that pregnancy did not adversely affect coupling in the long-term.

Localization of Arterial Bleeds Using Pulse Wave Reflections

ABSTRACT Introduction Localization of internal arterial bleeds is necessary for treatment in the battlefield. In this article, we describe a novel approach that utilizes pulse wave reflections generated by a bleed to locate it. Materials and Methods To demonstrate our approach, velocity and diameter waveforms in the presence of bleeds were simulated using the 1D wave propagation equations in a straight-vessel model of the human thoracic aorta. The simulated waveforms were then decomposed into forward and backward components using wave intensity analysis. Reflections arising from the bleed were identified from the decomposed waveforms. Results Reflection generated by the bleed introduced a new feature in the backward component, compared to the normal, no-bleed condition. The bleed location could be determined from the time delay between this reflection feature and the forward wave creating it, and the pulse wave velocity in the vessel. Conclusions The findings of this study could be utilized by ultrasound for hemorrhage localization.

Angiographic evidence of backward compression wave: systolic compression of septal perforators in a child with hypertrophic cardiomyopathy

Intracoronary wave intensity analysis in hypertrophic cardiomyopathy has shown a large backward compression wave due to compressive deformation of the intramyocardial coronary arteries in systole. The authors describe the angiographic evidence of this backward compression wave, which has not been described in this physiological context and can be a marker of poor prognosis.