Mitral Valve Atlas for Artificial Intelligence Predictions of MitraClip Intervention Outcomes

The future of valvular heart disease assessment and therapy

Valvular heart disease (VHD) is becoming more prevalent in an ageing population, leading to challenges in diagnosis and management. This two-part Series offers a comprehensive review of changing concepts in VHD, covering diagnosis, intervention timing, novel management strategies, and the current state of research. The first paper highlights the remarkable progress made in imaging and transcatheter techniques, effectively addressing the treatment paradox wherein populations at the highest risk of VHD often receive the least treatment. These advances have attracted the attention of clinicians, researchers, engineers, device manufacturers, and investors, leading to the exploration and proposal of treatment approaches grounded in pathophysiology and multidisciplinary strategies for VHD management. This Series paper focuses on innovations involving computational, pharmacological, and bioengineering approaches that are transforming the diagnosis and management of patients with VHD. Artificial intelligence and digital methods are enhancing screening, diagnosis, and planning procedures, and the integration of imaging and clinical data is improving the classification of VHD severity. The emergence of artificial intelligence techniques, including so-called digital twins-eg, computer-generated replicas of the heart-is aiding the development of new strategies for enhanced risk stratification, prognostication, and individualised therapeutic targeting. Various new molecular targets and novel pharmacological strategies are being developed, including multiomics-ie, analytical methods used to integrate complex biological big data to find novel pathways to halt the progression of VHD. In addition, efforts have been undertaken to engineer heart valve tissue and provide a living valve conduit capable of growth and biological integration. Overall, these advances emphasise the importance of early detection, personalised management, and cutting-edge interventions to optimise outcomes amid the evolving landscape of VHD. Although several challenges must be overcome, these breakthroughs represent opportunities to advance patient-centred investigations.

Patient-Specific Quantitative In-Vivo Assessment of Human Mitral Valve Leaflet Strain Before and After MitraClip Repair

PURPOSE

Mitral regurgitation (MR) is a highly prevalent and deadly cardiac disease characterized by improper mitral valve (MV) leaflet coaptation. Among the plethora of available treatment strategies, the MitraClip is an especially safe option, but optimizing its long-term efficacy remains an urgent challenge.

METHODS

We applied our noninvasive image-based strain computation pipeline [1] to intraoperative transesophageal echocardiography datasets taken from ten patients undergoing MitraClip repair, spanning a range of MR etiologies and MitraClip configurations. We then analyzed MV leaflet strains before and after MitraClip implementation to develop a better understanding of (1) the pre-operative state of human regurgitant MV, and (2) the MitraClip's impact on the MV leaflet deformations.

RESULTS

The MV pre-operative strain fields were highly variable, underscoring both the heterogeneity of the MR in the patient population and the need for patient-specific treatment approaches. Similarly, there were no consistent overall post-operative strain patterns, although the average A2 segment radial strain difference between pre- and post-operative states was consistently positive. In contrast, the post-operative strain fields were better correlated to their respective pre-operative strain fields than to the inter-patient post-operative strain fields. This quantitative result implies that the patient specific pre-operative state of the MV guides its post-operative deformation, which suggests that the post-operative state can be predicted using pre-operative data-derived modelling alone.

CONCLUSIONS

The pre-operative MV leaflet strain patterns varied considerably across the range of MR disease states and after MitraClip repair. Despite large inter-patient heterogeneity, the post-operative deformation appears principally dictated by the pre-operative deformation state. This novel finding suggests that though the variation in MR functional state and MitraClip-induced deformation were substantial, the post-operative state can be predicted from the pre-operative data alone. This study suggests that, with use of larger patient cohort and corresponding long-term outcomes, quantitative predictive factors of MitraClip durability can be identified.

Emerging Roles for Artificial Intelligence in Heart Failure Imaging

Artificial intelligence (AI) applications are expanding in cardiac imaging. AI research has shown promise in workflow optimization, disease diagnosis, and integration of clinical and imaging data to predict patient outcomes. The diagnostic and prognostic paradigm of heart failure is heavily reliant on cardiac imaging. As AI becomes increasingly validated and integrated into clinical practice, AI influence on heart failure management will grow. This review discusses areas of current research and potential clinical applications in AI as applied to heart failure cardiac imaging.

The evolution of technical prerequisites and local boundary conditions for optimization of mitral valve interventions-Emphasis on skills development and institutional risk performance

This viewpoint report describes how the evolution of transcatheter mitral valve intervention (TMVI) is influenced by lessons learned from three evolutionary tracks: (1) the development of treatment from mitral valve surgery (MVS) to transcutaneous procedures; (2) the evolution of biomedical engineering for research and development resulting in predictable and safe clinical use; (3) the adaptation to local conditions, impact of transcatheter aortic valve replacement (TAVR) experience and creation of infrastructure for skills development and risk management. Thanks to developments in computer science and biostatistics, an increasing number of reports regarding clinical safety and effectiveness is generated. A full toolbox of techniques, devices and support technology is now available, especially in surgery. There is no doubt that the injury associated with a minimally invasive access reduces perioperative risks, but it may affect the effectiveness of the treatment due to incomplete correction. Based on literature, solutions and performance standards are formulated with an emphasis in technology and positive outcome. Despite references to Heart Team decision making, boundary conditions such as hospital infrastructure, caseload, skills training and perioperative risk management remain underexposed. The role of Biomedical Engineering is exclusively defined by the Research and Development (R&D) cycle including the impact of human factor engineering (HFE). Feasibility studies generate estimations of strengths and safety limitations. Usability testing reveals user friendliness and safety margins of clinical use. Apart from a certification requirement, this information should have an impact on the definition of necessary skills levels and consequent required training. Physicians Preference Testing (PPT) and use of a biosimulator are recommended. The example of the interaction between two Amsterdam heart centers describes the evolution of a professional ecosystem that can facilitate innovation. Adaptation to local conditions in terms of infrastructure, referrals and reimbursement, appears essential for the evolution of a complete mitral valve disease management program. Efficacy of institutional risk management performance (IRMP) and sufficient team skills should be embedded in an appropriate infrastructure that enables scale and offers complete and safe solutions for mitral valve disease. The longstanding evolution of mitral valve therapies is the result of working devices embedded in an ecosystem focused on developing skills and effective risk management actions.

Machine learning used for simulation of MitraClip intervention: A proof-of-concept study

Introduction: 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.

Methods: 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.

Results: 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.

Discussion: 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.

Looking towards the future: patient-specific computational modeling to optimize outcomes for transcatheter mitral valve repair

Severe mitral valve regurgitation (MR) is a heart valve disease that progresses to end-stage congestive heart failure and death if left untreated. Surgical repair or replacement of the mitral valve (MV) remains the gold standard for treatment of severe MR, with repair techniques aiming to restore the native geometry of the MV. However, patients with extensive co-morbidities may be ineligible for surgical intervention. With the emergence of transcatheter MV repair (TMVR) treatment paradigms for MR will evolve. The longer-term outcomes of TMVR and its effectiveness compared to surgical repair remain unknown given the differing patient eligibility for either treatment at this time. Advances in computational modeling will elucidate answers to these questions, employing techniques such as finite element method and fluid structure interactions. Use of clinical imaging will permit patient-specific MV models to be created with high accuracy and replicate MV pathophysiology. It is anticipated that TMVR technology will gradually expand to treat lower-risk patient groups, thus pre-procedural computational modeling will play a crucial role guiding clinicians towards the optimal intervention. Additionally, concerted efforts to create MV models will establish atlases of pathologies and biomechanics profiles which could delineate which patient populations would best benefit from specific surgical vs. TMVR options. In this review, we describe recent literature on MV computational modeling, its relevance to MV repair techniques, and future directions for translational application of computational modeling for treatment of MR.