Detecting Aortic Valve Anomaly From Induced Murmurs: Insights From Computational Hemodynamic Models

Abstract

Patients who receive transcatheter aortic valve replacement are at risk for leaflet thrombosis-related complications, and can benefit from continuous, longitudinal monitoring of the prosthesis. Conventional angiography modalities are expensive, hospital-centric and either invasive or employ potentially nephrotoxic contrast agents, which preclude their routine use. Heart sounds have been long recognized to contain valuable information about individual valve function, but the skill of auscultation is in decline due to its heavy reliance on the physician’s proficiency leading to poor diagnostic repeatability. This subjectivity in diagnosis can be alleviated using machine learning techniques for anomaly detection. We present a computational and data-driven proof-of-concept analysis of a novel, auscultation-based technique for monitoring aortic valve, which is practical, non-invasive, and non-toxic. However, the underlying mechanisms leading to physiological and pathological heart sounds are not well-understood, which hinders development of such a technique. We first address this by performing direct numerical simulations of the complex interactions between turbulent blood flow in a canonical ascending aorta model and dynamic valve motion in 29 cases with healthy and stenotic valves. Using the turbulent pressure fluctuations on the aorta lumen boundary, we model the propagation of heart sounds, as elastic waves, through the patient’s thorax. The heart sound may be recorded on the epidermal surface using a stethoscope/phonocardiograph. This approach allows us to correlate instantaneous hemodynamic phenomena and valve motion with the acoustic response. From this dataset we extract “acoustic signatures” of healthy and stenotic valves based on principal components of the recorded sound. These signatures are used to train a linear discriminant classifier by maximizing correlation between recorded heart sounds and valve status. We demonstrate that this classifier is capable of accurate prospective detection of anomalous valve function and that the principal component-based signatures capture prominent audible features of heart sounds, which have been historically used by physicians for diagnosis. Further development of such technology can enable inexpensive, safe and patient-centric at-home monitoring, and can extend beyond transcatheter valves to surgical as well as native valves.

Changes in aorta hemodynamics in Left-Right Type 1 bicuspid aortic valve patients after replacement with bioprosthetic valves: An in-silico study

“…The solver has previously been successfully employed for direct numerical simulations of flow through stenosed aortic valves (AS) [44,45], mitral valves [46] and left ventricles [46,47], as well as several other biological flow systems [48,49]. The solver is coupled with a versatile reduced degree-of-freedom (rDOF) valve model [45, [50][51][52][53][54] which considerably simplifies the structural subsystem, reduces simulation time can be adapted to patient-specific annular morphology and accurately simulate various valvular pathologies. The model is used to simulate the response of severely stenotic BAVs and healthy TAVs to physiological blood flow for a set of patient-specific AAo models.…”

A 3D scaling law for supravalvular aortic stenosis suited for stethoscopic auscultations

“…Specifically, a low diagnostic accuracy was observed to be of international concern [25] . Generally, the underlying flow phenomena that lead to such murmurs are complex and not yet fully understood [26] . Moreover, it remains difficult with today's technology to perform heart murmur auscultations while simultaneously imaging a patient's hemodynamics, to clinically connect the two.…”

“…Nevertheless, connecting their underlying mechanisms remains elusive, due to their evolutionary complexity at the phenomenological scale. In-silico modeling was thus proposed as a potential solution by many researchers (e.g., Jiang et al [30] , Oshin et al [31] , Richez et al [32] , Morris et al [33] , and Bailoor et al [26] ). Such works illustrate the use of different computational methods to shed light on a variety of underlying hemodynamic and hemoacoustic mechanisms in the cardiovascular system.…”

“…The analysis of simulated bruits was elaborated on in several other computational studies (e.g., Tse et al [54] , Zhu et al [55] , and Bailoor et al [26] ). All these studies rely on the concept of the break frequency , i.e., a frequency that marks a sudden change in spectral power trends, consistent with in-vitro and clinical phonoangiography.…”

A Machine Anomalous Sound Detection Method Using the lMS Spectrogram and ES-MobileNetV3 Network

“…The anomalous sound detection (ASD) system, which relies on the perception, processing, and decision making of sound signals, is garnering increasing enthusiasm in the field of monitoring, owing to its advantages in safeguarding user privacy [1], real-time capabilities [2], and adaptability to complex environments [3]. Currently, anomalous sound detection technology finds widespread application in various fields, including animal husbandry [4,5], machine condition monitoring [6,7], medical surveillance [8,9], and many other aspects.…”