Hemodynamics-driven mathematical model of first and second heart sound generation.

NEW & NOTEWORTHY: To the best of our knowledge, this is the first hemodynamic-based heart sound generation model embedded in a complete real-time computational model of the cardiovascular system. Simulated heart sounds are similar to experimental and clinical measurements, both quantitatively and qualitatively. Our model can be used to investigate the relationships between heart sound acoustic features and hemodynamic factors/anatomical parameters.

The 'Digital Twin' to enable the vision of precision cardiology.

Providing therapies tailored to each patient is the vision of precision medicine, enabled by the increasing ability to capture extensive data about individual patients. In this position paper, we argue that the second enabling pillar towards this vision is the increasing power of computers and algorithms to learn, reason, and build the 'digital twin' of a patient. Computational models are boosting the capacity to draw diagnosis and prognosis, and future treatments will be tailored not only to current health status and data, but also to an accurate projection of the pathways to restore health by model predictions. The early steps of the digital twin in the area of cardiovascular medicine are reviewed in this article, together with a discussion of the challenges and opportunities ahead. We emphasize the synergies between mechanistic and statistical models in accelerating cardiovascular research and enabling the vision of precision medicine.

Machine learning-powered, device-embedded heart sound measurement can optimize AV delay in patients with CRT.

BACKGROUND: Continuous optimization of atrioventricular (AV) delay for cardiac resynchronization therapy (CRT) is mainly performed by electrical means. OBJECTIVE: The purpose of this study was to develop an estimation model of cardiac function that uses a piezoelectric microphone embedded in a pulse generator to guide CRT optimization. METHODS: Electrocardiogram, left ventricular pressure (LVP), and heart sounds were simultaneously collected during CRT device implantation procedures. A piezoelectric alarm transducer embedded in a modified CRT device facilitated recording of heart sounds in patients undergoing a pacing protocol with different AV delays. Machine learning (ML) was used to produce a decision-tree ensemble model capable of estimating absolute maximal LVP (LVPmax) and maximal rise of LVP (LVdP/dtmax) using 3 heart sound-based features. To gauge the applicability of ML in AV delay optimization, polynomial curves were fitted to measured and estimated values. RESULTS: In the data set of ∼30,000 heartbeats, ML indicated S1 amplitude, S2 amplitude, and S1 integral (S1 energy for LVdP/dtmax) as most prominent features for AV delay optimization. ML resulted in single-beat estimation precision for absolute values of LVPmax and LVdP/dtmax of 67% and 64%, respectively. For 20-30 beat averages, cross-correlation between measured and estimated LVPmax and LVdP/dtmax was 0.999 for both. The estimated optimal AV delays were not significantly different from those measured using invasive LVP (difference -5.6 ± 17.1 ms for LVPmax and +5.1 ± 6.7 ms for LVdP/dtmax). The difference in function at estimated and measured optimal AV delays was not statiscally significant (1 ± 3 mm Hg for LVPmax and 9 ± 57 mm Hg/s for LVdP/dtmax). CONCLUSION: Heart sound sensors embedded in a CRT device, powered by a ML algorithm, provide a reliable assessment of optimal AV delays and absolute LVPmax and LVdP/dtmax.

Heart sound-derived systolic time intervals for atrioventricular delay optimization in cardiac resynchronization therapy.

BACKGROUND: Phonocardiography (PCG) can be used to determine systolic time intervals (STIs) from ventricular pacing spike to the first heart sound (VS1) and from the first to the second heart sound (S1S2). OBJECTIVE: The purpose of this study was to investigate the relations between STIs and hemodynamics during atrioventricular (AV) delay optimization of biventricular pacing (BiVP) in animals and patients. METHODS: Five pigs with AV block underwent BiVP, while PCG was collected from an epicardial accelerometer. In 21 patients undergoing cardiac resynchronization therapy device implantation, PCG was recorded with a pulse generator-embedded microphone. Optimal AV delays derived from shortest VS1 and longest S1S2 were compared with AV delays derived from highest left ventricular pressure (LVP), maximal rate of rise in LVP, and stroke work. RESULTS: In pigs, VS1 and S1S2 predicted the AV delays with optimal hemodynamics (highest LVP, maximal rate of rise in LVP, and stroke work) by a median error of 2-28 ms, resulting in a median loss of <2% of pump function. In patients, VS1 and S1S2 predicted the optimal AV delay by errors of 32.5 and 37.5 ms, respectively, resulting in 0.2%-0.9% lower LVP and stroke work, which were reduced to 21 and 24 ms in 8 patients with a full-capture AV delay of >180 ms. CONCLUSION: During BiVP with varying AV delays, close relations exist between PCG-derived STIs and hemodynamic parameters. AV delays advised by PCG-derived STIs cause only a minimal loss of pump function compared with those based on invasive hemodynamic measurements. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01832493.

Hemodynamics-driven mathematical model of third heart sound generation.

The proto-diastolic third heart sound (S3) is observed in various hemodynamic conditions in both normal and diseased hearts. We propose a novel, one-degree of freedom mathematical model of mechanical vibrations of heart and blood that generates the third heart sound, implemented in a real-time model of the cardiovascular system (CircAdapt). To examine model functionality, S3 simulations were performed for conditions mimicking the normal heart as well as heart failure with preserved ejection fraction (HFpEF), atrioventricular valve regurgitation (AVR), atrioventricular valve stenosis (AVS) and septal shunts (SS). Simulated S3 showed both qualitative and quantitative agreements with measured S3 in terms of morphology, frequency, and timing. It was shown that ventricular mass, ventricular viscoelastic properties as well as inflow momentum play a key role in the generation of S3. The model indicated that irrespective of cardiac conditions, S3 vibrations are always generated, in both the left and right sides of the heart, albeit at different levels of audibility. S3 intensities increased in HFpEF, AVR and SS, but the changes of acoustic S3 features in AVS were not significant, as compared with the reference simulation. S3 loudness in all simulated conditions was proportional to the level of cardiac output and severity of cardiac conditions. In conclusion, our hemodynamics-driven mathematical model provides a fast and realistic simulation of S3 under various conditions which may be helpful to find new indicators for diagnosis and prognosis of cardiac diseases.

Left Ventricular Pressure Estimation Using Machine Learning-Based Heart Sound Classification.

Objective: A method to estimate absolute left ventricular (LV) pressure and its maximum rate of rise (LV dP/dtmax) from epicardial accelerometer data and machine learning is proposed. Methods: Five acute experiments were performed on pigs. Custom-made accelerometers were sutured epicardially onto the right ventricle, LV, and right atrium. Different pacing configurations and contractility modulations, using isoflurane and dobutamine infusions, were performed to create a wide variety of hemodynamic conditions. Automated beat-by-beat analysis was performed on the acceleration signals to evaluate amplitude, time, and energy-based features. For each sensing location, bootstrap aggregated classification tree ensembles were trained to estimate absolute maximum LV pressure (LVPmax) and LV dP/dtmax using amplitude, time, and energy-based features. After extraction of acceleration and pressure-based features, location specific, bootstrap aggregated classification ensembles were trained to estimate absolute values of LVPmax and its maximum rate of rise (LV dP/dtmax) from acceleration data. Results: With a dataset of over 6,000 beats, the algorithm narrowed the selection of 17 predefined features to the most suitable 3 for each sensor location. Validation tests showed the minimal estimation accuracies to be 93% and 86% for LVPmax at estimation intervals of 20 and 10 mmHg, respectively. Models estimating LV dP/dtmax achieved an accuracy of minimal 93 and 87% at estimation intervals of 100 and 200 mmHg/s, respectively. Accuracies were similar for all sensor locations used. Conclusion: Under pre-clinical conditions, the developed estimation method, employing epicardial accelerometers in conjunction with machine learning, can reliably estimate absolute LV pressure and its first derivative.

Second heart sound splitting as an indicator of interventricular mechanical dyssynchrony using a novel splitting detection algorithm.

Second heart sound (S2) splitting results from nonsimultaneous closures between aortic (A2) and pulmonic valves (P2) and may be used to detect timing differences (dyssynchrony) in relaxation between right (RV) and left ventricle (LV). However, overlap of A2 and P2 and the change in heart sound morphologies have complicated detection of the S2 splitting interval. This study introduces a novel S-transform amplitude ridge tracking (START) algorithm for estimating S2 splitting interval and investigates the relationship between S2 splitting and interventricular relaxation dyssynchrony (IRD). First, the START algorithm was validated in a simulated model of heart sound. It showed small errors (<5 ms) in estimating splitting intervals from 10 to 70 ms, with A2/P2 amplitude ratios from 0.2 to 5, and signal-to-noise ratios from 10 to 30 dB. Subsequently, the START algorithm was evaluated in a porcine model employing a wide range of paced RV-LV delays. IRD was quantified by the time difference between invasively measured LV and RV pressure downslopes. Between LV pre-excitation to RV pre-excitation, mean S2 splitting interval decreased from 47 ms to 23 ms (p < .001), accompanied by a decrease in mean IRD from 8 ms to -18 ms (p < .001). S2 splitting interval was significantly correlated with IRD in each experiment (p < .001). In conclusion, the START algorithm can accurately assess S2 splitting and may serve as a useful tool to assess interventricular dyssynchrony.