An understanding of left ventricle (LV) mechanics is fundamental for designing better preventive, diagnostic, and treatment strategies for improved heart function. Because of the costs of clinical ...and experimental studies to treat and understand heart function, respectively, in-silico models play an important role. Finite element (FE) models, which have been used to create in-silico LV models for different cardiac health and disease conditions, as well as cardiac device design, are time-consuming and require powerful computational resources, which limits their use when real-time results are needed. As an alternative, we sought to use deep learning (DL) for LV in-silico modeling. We used 80 four-chamber heart FE models for feed forward, as well as recurrent neural network (RNN) with long short-term memory (LSTM) models for LV pressure and volume. We used 120 LV-only FE models for training LV stress predictions. The active material properties of the myocardium and time were features for the LV pressure and volume training, and passive material properties and element centroid coordinates were features of the LV stress prediction models. For six test FE models, the DL error for LV volume was 1.599 ± 1.227 ml, and the error for pressure was 1.257 ± 0.488 mmHg; for 20 LV FE test examples, the mean absolute errors were, respectively, 0.179 ± 0.050 for myofiber, 0.049 ± 0.017 for cross-fiber, and 0.039 ± 0.011 kPa for shear stress. After training, the DL runtime was in the order of seconds whereas equivalent FE runtime was in the order of several hours (pressure and volume) or 20 min (stress). We conclude that using DL, LV in-silico simulations can be provided for applications requiring real-time results.
Mitral regurgitation (MR) is the most common type of valvular heart disease in patients over the age of 75 in the US. Despite the prevalence of mitral regurgitation in the elderly population, ...however, almost half of patients identified with moderate-severe MR are turned down for traditional open heart surgery due to frailty and other existing co-morbidities. MitraClip (MC) is a recent percutaneous approach to treat mitral regurgitation by placement of MC in the center of the mitral valve to reduce MR. There are currently no computational simulations to elucidate the role of MC on both the fluid and solid mechanics of the mitral valve. Here, we use the Smoothed Particle Hydrodynamics (SPH) approach to study various positional placements of the MC in the mitral valve and its impact on reducing MR. SPH is a particle based (meshless) approach that handles flow through narrow regions quite efficiently. Fluid and surrounding anatomical structure interactions is handled via contact and hence can be used for studying fluid-structure interaction problems such as blood flow with surrounding tissues/structure. This method is available as part of the Abaqus/Explicit solver. Regurgitation was initiated by removing targeted chordae tendineae that are attached to specified leaflets of the mitral valve and, subsequently, MC implants are placed in various locations, starting from the region near where the chordae tendineae were removed and moving away from the location towards the center of the valve. The MC implant location closest to where the chordae tendineae were removed showed the least amount of residual MR post-clip implantation amongst all other locations of MC implant considered. These findings have important implications for strategic placement of the MC depending on the etiology of MR to optimize clinical outcome.
Predictive computation models offer the potential to uncover the mechanisms of treatments whose actions cannot be easily determined by experimental or imaging techniques. This is particularly ...relevant for investigating left ventricular mechanical assistance, a therapy for end-stage heart failure, which is increasingly used as more than just a bridge-to-transplant therapy. The high incidence of right ventricular failure following left ventricular assistance reflects an undesired consequence of treatment, which has been hypothesized to be related to the mechanical interdependence between the two ventricles. To investigate the implication of this interdependence specifically in the setting of left ventricular assistance device (LVAD) support, we introduce a patient-specific finite-element model of dilated chronic heart failure. The model geometry and material parameters were calibrated using patient-specific clinical data, producing a mechanical surrogate of the failing
heart that models its dynamic strain and stress throughout the cardiac cycle. The model of the heart was coupled to lumped-parameter circulatory systems to simulate realistic ventricular loading conditions. Finally, the impact of ventricular assistance was investigated by incorporating a pump with pressure-flow characteristics of an LVAD (HeartMate II™ operating between 8 and 12 k RPM) in parallel to the left ventricle. This allowed us to investigate the mechanical impact of acute left ventricular assistance at multiple operating-speeds on right ventricular mechanics and septal wall motion. Our findings show that left ventricular assistance reduces myofiber stress in the left ventricle and, to a lesser extent, right ventricle free wall, while increasing leftward septal-shift with increased operating-speeds. These effects were achieved with secondary, potentially negative effects on the interventricular septum which showed that support from LVADs, introduces unnatural bending of the septum and with it, increased localized stress regions. Left ventricular assistance unloads the left ventricle significantly and shifts the right ventricular pressure-volume-loop toward larger volumes and higher pressures; a consequence of left-to-right ventricular interactions and a leftward septal shift. The methods and results described in the present study are a meaningful advancement of computational efforts to investigate heart-failure therapies
and illustrate the potential of computational models to aid understanding of complex mechanical and hemodynamic effects of new therapies.
Severe mitral regurgitation (MR) is a mitral valve disease that can lead to lifethreatening complications. MitraClip (MC) therapy is a percutaneous solution for patients who cannot tolerate surgical ...solutions. In MC therapy, a clip is implanted in the heart to reduce MR. To achieve optimal MC therapy, the cardiologist needs to foresee the outcomes of different scenarios for MC implantation, including the location of the MC. Although finite element (FE) modeling can simulate the outcomes of different MC scenarios, it is not suitable for clinical usage because it requires several hours to complete.
In this paper, we used machine learning (ML) to predict the outcomes of MC therapy in less than 1 s. Two ML algorithms were used: XGBoost, which is a decision tree model, and a feed-forward deep learning (DL) model. The MC location, the geometrical attributes of the models and baseline stress and MR were the features of the ML models, and the predictions were performed for MR and maximum von Mises stress in the leaflets. The parameters of the ML models were determined to achieve the minimum errors obtained by applying the ML models on the validation set.
The results for the test set (not used during training) showed relative agreement between ML predictions and ground truth FE predictions. The accuracy of the XGBoost models were better than DL models. Mean absolute percentage error (MAPE) for the XGBoost predictions were 0.115 and 0.231, and the MAPE for DL predictions were 0.154 and 0.310, for MR and stress, respectively.
The ML models reduced the FE runtime from 6 hours (on average) to less than 1 s. The accuracy of ML models can be increased by increasing the dataset size. The results of this study have important implications for improving the outcomes of MC therapy by providing information about the outcomes of MC implantation in real-time.
The pathophysiological mechanisms underlying preserved left ventricular (LV) ejection fraction (EF) in patients with heart failure and preserved ejection fraction (HFpEF) remain incompletely ...understood. We hypothesized that transmural variations in myofiber contractility with existence of subendocardial dysfunction and compensatory increased subepicardial contractility may underlie preservation of LVEF in patients with HFpEF. We quantified alterations in myocardial function in a mathematical model of the human LV that is based on the finite element method. The fiber-reinforced material formulation of the myocardium included passive and active properties. The passive material properties were determined such that the diastolic pressure-volume behavior of the LV was similar to that shown in published clinical studies of pressure-volume curves. To examine changes in active properties, we considered six scenarios: (1) normal properties throughout the LV wall; (2) decreased myocardial contractility in the subendocardium; (3) increased myocardial contractility in the subepicardium; (4) myocardial contractility decreased equally in all layers, (5) myocardial contractility decreased in the midmyocardium and subepicardium, (6) myocardial contractility decreased in the subepicardium. Our results indicate that decreased subendocardial contractility reduced LVEF from 53.2 to 40.5%. Increased contractility in the subepicardium recovered LVEF from 40.5 to 53.2%. Decreased contractility transmurally reduced LVEF and could not be recovered if subepicardial and midmyocardial contractility remained depressed. The computational results simulating the effects of transmural alterations in the ventricular tissue replicate the phenotypic patterns of LV dysfunction observed in clinical practice. In particular, data for LVEF, strain and displacement are consistent with previous clinical observations in patients with HFpEF, and substantiate the hypothesis that increased subepicardial contractility may compensate for subendocardial dysfunction and play a vital role in maintaining LVEF.
The goal of this paper was to provide a real-time left ventricular (LV) mechanics simulator using machine learning (ML). Finite element (FE) simulations were conducted for the LV with different ...material properties to obtain a training set. A hyperelastic fiber-reinforced material model was used to describe the passive behavior of the myocardium during diastole. The active behavior of the heart resulting from myofiber contractions was added to the passive tissue during systole. The active and passive properties govern the LV constitutive equation. These mechanical properties were altered using optimal Latin hypercube design of experiments to obtain training FE models with varied active properties (volume and pressure predictions) and varied passive properties (stress predictions). For prediction of LV pressures, we used eXtreme Gradient Boosting (XGboost) and Cubist, and XGBoost was used for predictions of LV pressures, volumes as well as LV stresses. The LV pressure and volume results obtained from ML were similar to FE computations. The ML results could capture the shape of LV pressure as well as LV pressure-volume loops. The results predicted by Cubist were smoother than those from XGBoost. The mean absolute errors were as follows: XGBoost volume: 1.734 ± 0.584 ml, XGBoost pressure: 1.544 ± 0.298 mmHg, Cubist volume: 1.495 ± 0.260 ml, Cubist pressure: 1.623 ± 0.191 mmHg, myofiber stress: 0.334 ± 0.228 kPa, cross myofiber stress: 0.075 ± 0.024 kPa, and shear stress: 0.050 ± 0.032 kPa. The simulation results show ML can predict LV mechanics much faster than the FE method. The ML model can be used as a tool to predict LV behavior. Training of our ML model based on a large group of subjects can improve its predictability for real world applications.
Severe mitral regurgitation (MR) is a cardiac disease that can lead to fatal consequences. MitraClip (MC) intervention is a percutaneous procedure whereby the mitral valve (MV) leaflets are connected ...along the edge using MCs. The outcomes of the MC intervention are not known in advance, i.e., the outcomes are quite variable. Artificial intelligence (AI) can be used to guide the cardiologist in selecting optimal MC scenarios. In this study, we describe an atlas of shapes as well as different scenarios for MC implantation for such an AI analysis. We generated the MV geometrical data from three different sources. First, the patients' 3-dimensional echo images were used. The pixel data from six key points were obtained from three views of the echo images. Using PyGem, an open-source morphing library in Python, these coordinates were used to create the geometry by morphing a template geometry. Second, the dimensions of the MV, from the literature were used to create data. Third, we used machine learning methods, principal component analysis, and generative adversarial networks to generate more shapes. We used the finite element (FE) software ABAQUS to simulate smoothed particle hydrodynamics in different scenarios for MC intervention. The MR and stresses in the leaflets were post-processed. Our physics-based FE models simulated the outcomes of MC intervention for different scenarios. The MR and stresses in the leaflets were computed by the FE models for a single clip at different locations as well as two and three clips. Results from FE simulations showed that the location and number of MCs affect subsequent residual MR, and that leaflet stresses do not follow a simple pattern. Furthermore, FE models need several hours to provide the results, and they are not applicable for clinical usage where the predicted outcomes of MC therapy are needed in real-time. In this study, we generated the required dataset for the AI models which can provide the results in a matter of seconds.
Abstract only
Introduction
anorectal function is complex. Assessment of anorectal physiology and diagnostics for anorectal disorders face problems due to inconsistent results with different ...technologies and huge inter‐subject variability. This limits development of therapeutic strategies. Devices used to evaluate anorectal function include Fecobionics, EndoFLIP, Anorectal manometry and Ultrasound. The goal of this study was to find correlation in patient characteristics based on data obtained with these measurements.
Methods
the four tests mentioned above were used to evaluate anorectal function in normal subjects and patients with fecal incontinence (FI) or constipation. The data were cleaned by replacing unknown values with the average of measurements, and they were standardized. In total there were 90 measurements and 35 features. The collected data included sex, age (60.0±14.3 years), weight (59.0±9.0 kg), height (1.6±0.1m), FI score (16.8±12.8 with normal values =< 5), constipation score (6.5±4.4 with normal values =< 8), anorectal maximum pressure (46.9±10.1 mmHg), resting pressure (61±25.8 mmHg), expel time (46.3±63.2s), and absence or presentation of rectoanal inhibitory reflex (RAIR), etc. Principal component analysis (PCA) was used to find the contribution of each reduced feature in variance of data and the correlation between features, utilizing singular value decomposition. We used Python and its libraries.
Results and Discussion
the variance in data was mostly explained using 25 PCs. The percentage of cumulative variance explained by up to the 1
th
, 25
th
and 35
th
PC was 15%, 98% and 100%, respectively. The PCs provided insights about the relation between features. For example, according to the first PC the maximum pressure, resting pressure, maximum tolerance and age were positively correlated. Also, weight, absence of RAIR and constipation score were positively correlated. However, the latter group of parameters were negatively correlated to the former group. Because the number of patients used in this study were limited results should be interpreted with caution. However, more data are currently being collected, which will be included in our analysis. We plan to extend our analysis to cluster patient data using machine learning.
Conclusions
in this study we analyzed data gathered in anorectal evaluation using PCA. Data obtained provided insights about relation between characteristics of the patients. Analysis of data obtained in anorectal evaluation will help to better understand factors involved in anorectal and other gastrointestinal disorders, which could lead to better preventive, diagnostic and therapeutic strategies.
Support or Funding Information
NIH SPARC funding
Acknowledgements
NIH SPARC funding.
Abstract only
Introduction
Lymphatic system is an important part of the circulation system. Assessment of alterations in lymphatic flow in diseased conditions could lead to better treatment ...strategies. In vivo determination of the flow and pressure in the lymphatic vessels can be challenging. Computational modeling can be used to simulate the lymph flow in silico. The goal of this study was to develop a platform whereby the lymphatic flow in a network of vessels can be simulated. The mathematical model was based on animal experiments.
Methods
We conducted swine experiments whereby the connections between lymphatic vessels in the hind limb were visualized. The length and dimeter of the vessels were also determined from animal experiments. Finite element (FE) method was used to discretize one dimensional lymph flow equations. The lymph vessels were considered as extensible cylinders where the relation between pressure drop and flow was based on experimental data. Also, for valves, the relation between pressure drop and flow was based on experimental data. At junctions we assumed pressure drop is negligible and the flow in the proximal vessel is equal to the sum of flow in distal vessels. The inflow velocities at the distal vessels and the pressure at the proximal vessel were used as boundary conditions. We conducted a time‐dependent analysis where the initial conditions were obtained from a steady state solution.
Results and Discussion
When number of the elements and time steps were increased the solution converged. For a steady‐state rigid vessel, the results were in agreement with the Poiseuille's law. Also, for a branch with rigid vessels the results were in agreement with analytical solutions. For models with extensible vessels, the computed pressure and velocities corresponded to the experimental data. The platform was applicable to junctions with different number of vessels. Because the dimension of lymphatic vessels is noticeably small, measurement of diameters, flow and pressure is challenging, which limits the data that can be fed to the numerical model. As a future direction, this platform could be used to predict flow characteristics in different disease conditions, and possibility to perform more efficient amputations.
Conclusion
Computational modeling was utilized to simulate flow and pressure in lymphatic vessels. There are limitations in measurements of flow characteristics in the lymphatic vessels. Our computational framework could be used to better understanding of lymphatic system functionality in health and disease, and to simulate outcomes which are difficult to determine experimentally.
Support or Funding Information
This work was supported by the Office of the Assistant Secretary of Defense for Health Affairs, through the Defense Medical Research and Development Program under Award No. W81XWH‐19‐1‐0087. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the Department of Defense