Towards patient-specific modeling of mitral valve repair: 3D transesophageal echocardiography-derived parameter estimation

A geometry-based finite element tool for evaluating mitral valve biomechanics

Mitral valve function depends on its complex geometry and tissue health, with alterations in shape and tissue response affecting the long-term restorarion of function. Previous computational frameworks for biomechanical assessment are mostly based on patient-specific geometries; however, these are not flexible enough to yield a variety of models and assess mitral closure for individually tuned morphological parameters or material property representations. This study details the finite element approach implemented in our previously developed toolbox to assess mitral valve biomechanics and showcases its flexibility through the generation and biomechanical evaluation of different models. A healthy valve geometry was generated and its computational predictions for biomechanics validated against data in the literature. Moreover, two mitral valve models including geometric alterations associated with disease were generated and analysed. The healthy mitral valve model yielded biomechanical predictions in terms of valve closure dynamics, leaflet stresses and papillary muscle and chordae forces comparable to previous computational and experimental studies. Mitral valve function was compromised in geometries representing disease, expressed by the presence of regurgitating areas, elevated stress on the leaflets and unbalanced subvalvular apparatus forces. This showcases the flexibility of the toolbox concerning the generation of a range of mitral valve models with varying geometric definitions and material properties and the evaluation of their biomechanics.

Clinical Impact of Computational Heart Valve Models

This paper provides a review of engineering applications and computational methods used to analyze the dynamics of heart valve closures in healthy and diseased states. Computational methods are a cost-effective tool that can be used to evaluate the flow parameters of heart valves. Valve repair and replacement have long-term stability and biocompatibility issues, highlighting the need for a more robust method for resolving valvular disease. For example, while fluid-structure interaction analyses are still scarcely utilized to study aortic valves, computational fluid dynamics is used to assess the effect of different aortic valve morphologies on velocity profiles, flow patterns, helicity, wall shear stress, and oscillatory shear index in the thoracic aorta. It has been analyzed that computational flow dynamic analyses can be integrated with other methods to create a superior, more compatible method of understanding risk and compatibility.

Segmentation of Tricuspid Valve Leaflets From Transthoracic 3D Echocardiograms of Children With Hypoplastic Left Heart Syndrome Using Deep Learning

Hypoplastic left heart syndrome (HLHS) is a severe congenital heart defect in which the right ventricle and associated tricuspid valve (TV) alone support the circulation. TV failure is thus associated with heart failure, and the outcome of TV valve repair are currently poor. 3D echocardiography (3DE) can generate high-quality images of the valve, but segmentation is necessary for precise modeling and quantification. There is currently no robust methodology for rapid TV segmentation, limiting the clinical application of these technologies to this challenging population. We utilized a Fully Convolutional Network (FCN) to segment tricuspid valves from transthoracic 3DE. We trained on 133 3DE image-segmentation pairs and validated on 28 images. We then assessed the effect of varying inputs to the FCN using Mean Boundary Distance (MBD) and Dice Similarity Coefficient (DSC). The FCN with the input of an annular curve achieved a median DSC of 0.86 [IQR: 0.81-0.88] and MBD of 0.35 [0.23-0.4] mm for the merged segmentation and an average DSC of 0.77 [0.73-0.81] and MBD of 0.6 [0.44-0.74] mm for individual TV leaflet segmentation. The addition of commissural landmarks improved individual leaflet segmentation accuracy to an MBD of 0.38 [0.3-0.46] mm. FCN-based segmentation of the tricuspid valve from transthoracic 3DE is feasible and accurate. The addition of an annular curve and commissural landmarks improved the quality of the segmentations with MBD and DSC within the range of human inter-user variability. Fast and accurate FCN-based segmentation of the tricuspid valve in HLHS may enable rapid modeling and quantification, which in the future may inform surgical planning. We are now working to deploy this network for public use.

Artificial intelligence applied to support medical decisions for the automatic analysis of echocardiogram images: A systematic review

The echocardiogram is a test that is widely used in Heart Disease Diagnoses. However, its analysis is largely dependent on the physician's experience. In this regard, artificial intelligence has become an essential technology to assist physicians. This study is a Systematic Literature Review (SLR) of primary state-of-the-art studies that used Artificial Intelligence (AI) techniques to automate echocardiogram analyses. Searches on the leading scientific article indexing platforms using a search string returned approximately 1400 articles. After applying the inclusion and exclusion criteria, 118 articles were selected to compose the detailed SLR. This SLR presents a thorough investigation of AI applied to support medical decisions for the main types of echocardiogram (Transthoracic, Transesophageal, Doppler, Stress, and Fetal). The article's data extraction indicated that the primary research interest of the studies comprised four groups: 1) Improvement of image quality; 2) identification of the cardiac window vision plane; 3) quantification and analysis of cardiac functions, and; 4) detection and classification of cardiac diseases. The articles were categorized and grouped to show the main contributions of the literature to each type of ECHO. The results indicate that the Deep Learning (DL) methods presented the best results for the detection and segmentation of the heart walls, right and left atrium and ventricles, and classification of heart diseases using images/videos obtained by echocardiography. The models that used Convolutional Neural Network (CNN) and its variations showed the best results for all groups. The evidence produced by the results presented in the tabulation of the studies indicates that the DL contributed significantly to advances in echocardiogram automated analysis processes. Although several solutions were presented regarding the automated analysis of ECHO, this area of research still has great potential for further studies to improve the accuracy of results already known in the literature.

Applications of artificial intelligence in cardiovascular imaging

Research into artificial intelligence (AI) has made tremendous progress over the past decade. In particular, the AI-powered analysis of images and signals has reached human-level performance in many applications owing to the efficiency of modern machine learning methods, in particular deep learning using convolutional neural networks. Research into the application of AI to medical imaging is now very active, especially in the field of cardiovascular imaging because of the challenges associated with acquiring and analysing images of this dynamic organ. In this Review, we discuss the clinical questions in cardiovascular imaging that AI can be used to address and the principal methodological AI approaches that have been developed to solve the related image analysis problems. Some approaches are purely data-driven and rely mainly on statistical associations, whereas others integrate anatomical and physiological information through additional statistical, geometric and biophysical models of the human heart. In a structured manner, we provide representative examples of each of these approaches, with particular attention to the underlying computational imaging challenges. Finally, we discuss the remaining limitations of AI approaches in cardiovascular imaging (such as generalizability and explainability) and how they can be overcome.

3D Printing, Computational Modeling, and Artificial Intelligence for Structural Heart Disease

Structural heart disease (SHD) is a new field within cardiovascular medicine. Traditional imaging modalities fall short in supporting the needs of SHD interventions, as they have been constructed around the concept of disease diagnosis. SHD interventions disrupt traditional concepts of imaging in requiring imaging to plan, simulate, and predict intraprocedural outcomes. In transcatheter SHD interventions, the absence of a gold-standard open cavity surgical field deprives physicians of the opportunity for tactile feedback and visual confirmation of cardiac anatomy. Hence, dependency on imaging in periprocedural guidance has led to evolution of a new generation of procedural skillsets, concept of a visual field, and technologies in the periprocedural planning period to accelerate preclinical device development, physician, and patient education. Adaptation of 3-dimensional (3D) printing in clinical care and procedural planning has demonstrated a reduction in early-operator learning curve for transcatheter interventions. Integration of computation modeling to 3D printing has accelerated research and development understanding of fluid mechanics within device testing. Application of 3D printing, computational modeling, and ultimately incorporation of artificial intelligence is changing the landscape of physician training and delivery of patient-centric care. Transcatheter structural heart interventions are requiring in-depth periprocedural understanding of cardiac pathophysiology and device interactions not afforded by traditional imaging metrics.

Geometric description for the anatomy of the mitral valve: A review

The mitral valve is a complex anatomical structure whose physiological functioning relies on the biomechanical properties and structural integrity of its components. Their compromise can lead to mitral valve dysfunction, associated with morbidity and mortality. Therefore, a review on the morphometry of the mitral valve is crucial, more specifically on the importance of valve dimensions and shape for its function. This review initially provides a brief background on the anatomy and physiology of the mitral valve, followed by an analysis of the morphological information available. A characterisation of mathematical descriptions of several parts of the valve is performed and the impact of different dimensions and shape changes in disease is then outlined. Finally, a section regarding future directions and recommendations for the use of morphometric information in clinical analysis of the mitral valve is presented.

Mitral valve flattening and parameter mapping for patient-specific valve diagnosis

Purpose

Intensive planning and analysis from echocardiography are a crucial step before reconstructive surgeries are applied to malfunctioning mitral valves. Volume visualizations of echocardiographic data are often used in clinical routine. However, they lack a clear visualization of the crucial factors for decision making.

Methods

We build upon patient-specific mitral valve surface models segmented from echocardiography that represent the valve’s geometry, but suffer from self-occlusions due to complex 3D shape. We transfer these to 2D maps by unfolding their geometry, resulting in a novel 2D representation that maintains anatomical resemblance to the 3D geometry. It can be visualized together with color mappings and presented to physicians to diagnose the pathology in one gaze without the need for further scene interaction. Furthermore, it facilitates the computation of a Pathology Score, which can be used for diagnosis support.

Results

Quality and effectiveness of the proposed methods were evaluated through a user survey conducted with domain experts. We assessed pathology detection accuracy using 3D valve models in comparison with the novel visualizations. Classification accuracy increased by 5.3% across all tested valves and by 10.0% for prolapsed valves. Further, the participants’ understanding of the relation between 3D and 2D views was evaluated. The Pathology Score is found to have potential to support discriminating pathologic valves from normal valves.

Conclusions

In summary, our survey shows that pathology detection can be improved in comparison with simple 3D surface visualizations of the mitral valve. The correspondence between the 2D and 3D representations is comprehensible, and color-coded pathophysiological magnitudes further support the clinical assessment.

Temporal Views of Flattened Mitral Valve Geometries

The mitral valve, one of the four valves in the human heart, controls the bloodflow between the left atrium and ventricle and may suffer from various pathologies. Malfunctioning valves can be treated by reconstructive surgeries, which have to be carefully planned and evaluated. While current research focuses on the modeling and segmentation of the valve, we base our work on existing segmentations of patient-specific mitral valves, that are also time-resolved ( 3D+t) over the cardiac cycle. The interpretation of the data can be ambiguous, due to the complex surface of the valve and multiple time steps. We therefore propose a software prototype to analyze such 3D+t data, by extracting pathophysiological parameters and presenting them via dimensionally reduced visualizations. For this, we rely on an existing algorithm to unroll the convoluted valve surface towards a flattened 2D representation. In this paper, we show that the 3D+t data can be transferred to 3D or 2D representations in a way that allows the domain expert to faithfully grasp important aspects of the cardiac cycle. In this course, we not only consider common pathophysiological parameters, but also introduce new observations that are derived from landmarks within the segmentation model. Our analysis techniques were developed in collaboration with domain experts and a survey showed that the insights have the potential to support mitral valve diagnosis and the comparison of the pre- and post-operative condition of a patient.

A population-specific material model for sagittal craniosynostosis to predict surgical shape outcomes

Sagittal craniosynostosis consists of premature fusion (ossification) of the sagittal suture during infancy, resulting in head deformity and brain growth restriction. Spring-assisted cranioplasty (SAC) entails skull incisions to free the fused suture and insertion of two springs (metallic distractors) to promote cranial reshaping. Although safe and effective, SAC outcomes remain uncertain. We aimed hereby to obtain and validate a skull material model for SAC outcome prediction. Computed tomography data relative to 18 patients were processed to simulate surgical cuts and spring location. A rescaling model for age matching was created using retrospective data and validated. Design of experiments was used to assess the effect of different material property parameters on the model output. Subsequent material optimization-using retrospective clinical spring measurements-was performed for nine patients. A population-derived material model was obtained and applied to the whole population. Results showed that bone Young's modulus and relaxation modulus had the largest effect on the model predictions: the use of the population-derived material model had a negligible effect on improving the prediction of on-table opening while significantly improved the prediction of spring kinematics at follow-up. The model was validated using on-table 3D scans for nine patients: the predicted head shape approximated within 2 mm the 3D scan model in 80% of the surface points, in 8 out of 9 patients. The accuracy and reliability of the developed computational model of SAC were increased using population data: this tool is now ready for prospective clinical application.

Mitral Regurgitation Severity Analysis Based on Features and Optimal HE (OHE) with Quantification using PISA Method

Heart disease is the foremost reason for death and also the main source of incapability in the developed nations. Mitral regurgitation (MR) is a typical heart disease that does not bring about manifestations until its end position. In view of the hidden etiologies of heart distress, functional MR can be partitioned into two subgroups, ischemic and no ischemic MR. A procedure is progressed for jet area separation and quantification in MR evaluation in arithmetical expressions. Thus, a strategy that depends on echocardiography recordings, image processing methods, and artificial intelligence could be useful for clinicians, particularly in marginal cases. In this research paper, MR segmentation is analyzed by the optimal histogram equalization (OHE) system used to segment the jet area. For a better execution of the work, threshold in HE was improved with the help of the krill herd optimization (KHO) strategy. With the MR quantification procedure, this segmented jet area was supported by the proximal isovelocity surface area (PISA); in this procedure, a few parameters in the segmentation were evaluated. From the results, this proposed methodology accomplishes better accuracy in the segmented and quantification method in contrast with the existing examination.

Estimation of in vivo constitutive parameters of the aortic wall using a machine learning approach

The patient-specific biomechanical analysis of the aorta requires the quantification of the in vivo mechanical properties of individual patients. Current inverse approaches have attempted to estimate the nonlinear, anisotropic material parameters from in vivo image data using certain optimization schemes. However, since such inverse methods are dependent on iterative nonlinear optimization, these methods are highly computation-intensive. A potential paradigm-changing solution to the bottleneck associated with patient-specific computational modeling is to incorporate machine learning (ML) algorithms to expedite the procedure of in vivo material parameter identification. In this paper, we developed an ML-based approach to estimate the material parameters from three-dimensional aorta geometries obtained at two different blood pressure (i.e., systolic and diastolic) levels. The nonlinear relationship between the two loaded shapes and the constitutive parameters are established by an ML-model, which was trained and tested using finite element (FE) simulation datasets. Cross-validations were used to adjust the ML-model structure on a training/validation dataset. The accuracy of the ML-model was examined using a testing dataset.

Development of a Functionally Equivalent Model of the Mitral Valve Chordae Tendineae Through Topology Optimization

Ischemic mitral regurgitation (IMR) is a currently prevalent disease in the US that is projected to become increasingly common as the aging population grows. In recent years, image-based simulations of mitral valve (MV) function have improved significantly, providing new tools to refine IMR treatment. However, clinical implementation of MV simulations has long been hindered as the in vivo MV chordae tendineae (MVCT) geometry cannot be captured with sufficient fidelity for computational modeling. In the current study, we addressed this challenge by developing a method to produce functionally equivalent MVCT models that can be built from the image-based MV leaflet geometry alone. We began our analysis using extant micron-resolution 3D imaging datasets to first build anatomically accurate MV models. We then systematically simplified the native MVCT structure to generate a series of synthetic models by consecutively removing key anatomic features, such as the thickness variations, branching patterns, and chordal origin distributions. In addition, through topology optimization, we identified the minimal structural complexity required to capture the native MVCT behavior. To assess the performance and predictive power of each synthetic model, we analyzed their performance by comparing the mismatch in simulated MV closed shape, as well as the strain and stress tensors, to ground-truth MV models. Interestingly, our results revealed a substantial redundancy in the anatomic structure of native chordal anatomy. We showed that the closing behavior of complete MV apparatus under normal, diseased, and surgically repaired scenarios can be faithfully replicated by a functionally equivalent MVCT model comprised of two representative papillary muscle heads, single strand chords, and a uniform insertion distribution with a density of 15 insertions/cm2. Hence, even though the complete sub-valvular structure is mostly missing in in vivo MV images, we believe our approach will allow for the development of patient-specific complete MV models for surgical repair planning.

Computational Pre-surgical Planning of Arterial Patch Reconstruction: Parametric Limits and In Vitro Validation

Surgical treatment of congenital heart disease (CHD) involves complex vascular reconstructions utilizing artificial and native surgical materials. A successful surgical reconstruction achieves an optimal hemodynamic profile through the graft in spite of the complex post-operative vessel growth pattern and the altered pressure loading. This paper proposes a new in silico patient-specific pre-surgical planning framework for patch reconstruction and investigates its computational feasibility. The proposed protocol is applied to the patch repair of main pulmonary artery (MPA) stenosis in the Tetralogy of Fallot CHD template. The effects of stenosis grade, the three-dimensional (3D) shape of the surgical incision and material properties of the artificial patch are investigated. The release of residual stresses due to the surgical incision and the extra opening of the incision gap for patch implantation are simulated through a quasi-static finite-element vascular model with shell elements. Implantation of different unloaded patch shapes is simulated. The patched PA configuration is pressurized to the physiological post-operative blood pressure levels of 25 and 45 mmHg and the consequent post-operative stress distributions and patched artery shapes are computed. Stress–strain data obtained in-house, through the biaxial tensile tests for the mechanical properties of common surgical patch materials, Dacron, Polytetrafluoroethylene, human pericardium and porcine xenopericardium, are employed to represent the mechanical behavior of the patch material. Finite-element model is experimentally validated through the actual patch surgery reconstructions performed on the 3D printed anatomical stenosis replicas. The post-operative recovery of the initially narrowed lumen area and post-op tortuosity are quantified for all modeled cases. A computational fluid dynamics solver is used to evaluate post-operative pressure drop through the patch-reconstructed outflow tract. According to our findings, the shorter incisions made at the throat result in relatively low local peak stress values compared to other patch design alternatives. Longer cut and double patch cases are the most effective in repairing the initial stenosis level. After the patch insertion, the pressure drop in the artery due to blood flow decreases from 9.8 to 1.35 mmHg in the conventional surgical configuration. These results are in line with the clinical experience where a pressure gradient at or above 50 mmHg through the MPA can be an indication to intervene. The main strength of the proposed pre-surgical planning framework is its capability to predict the intra-operative and post-operative 3D vascular shape changes due to intramural pressure, cut length and configuration, for both artificial and native patch materials.

Estimation of in vivo mechanical properties of the aortic wall: A multi-resolution direct search approach

The patient-specific biomechanical analysis of the aorta requires in vivo mechanical properties of individual patients. Existing approaches for estimating in vivo material properties often demand high computational cost and mesh correspondence of the aortic wall between different cardiac phases. In this paper, we propose a novel multi-resolution direct search (MRDS) approach for estimation of the nonlinear, anisotropic constitutive parameters of the aortic wall. Based on the finite element (FE) updating scheme, the MRDS approach consists of the following three steps: (1) representing constitutive parameters with multiple resolutions using principal component analysis (PCA), (2) building links between the discretized PCA spaces at different resolutions, and (3) searching the PCA spaces in a 'coarse to fine' fashion following the links. The estimation of material parameters is achieved by minimizing a node-to-surface error function, which does not need mesh correspondence. The method was validated through a numerical experiment by using the in vivo data from a patient with ascending thoracic aortic aneurysm (ATAA), the results show that the number of FE iterations was significantly reduced compared to previous methods. The approach was also applied to the in vivo CT data from an aged healthy human patient, and using the estimated material parameters, the FE-computed geometry was well matched with the image-derived geometry. This novel MRDS approach may facilitate the personalized biomechanical analysis of aortic tissues, such as the rupture risk analysis of ATAA, which requires fast feedback to clinicians.

A new inverse method for estimation of in vivo mechanical properties of the aortic wall

The aortic wall is always loaded in vivo, which makes it challenging to estimate the material parameters of its nonlinear, anisotropic constitutive equation from in vivo image data. Previous approaches largely relied on either computationally expensive finite element models or simplifications of the geometry or material models. In this study, we investigated a new inverse method based on aortic wall stress computation. This approach consists of the following two steps: (1) computing an "almost true" stress field from the in vivo geometries and loading conditions, (2) building an objective function based on the "almost true" stress fields, constitutive equations and deformation relations, and estimating the material parameters by minimizing the objective function. The method was validated through numerical experiments by using the in vivo data from four ascending aortic aneurysm (AsAA) patients. The results demonstrated that the method is computationally efficient. This novel approach may facilitate the personalized biomechanical analysis of aortic tissues in clinical applications, such as in the rupture risk analysis of ascending aortic aneurysms.