Neural-Network-Based Diagnosis Using 3-Dimensional Myocardial Architecture and Deformation: Demonstration for the Differentiation of Hypertrophic Cardiomyopathy

Artificial intelligence for left ventricular hypertrophy detection and differentiation on echocardiography, cardiac magnetic resonance and cardiac computed tomography: A systematic review

AIMS

Left ventricular hypertrophy (LVH) is a common clinical finding associated with adverse cardiovascular outcomes. Once LVH is diagnosed, defining its cause has crucial clinical implications. Artificial intelligence (AI) may allow significant progress in the automated detection of LVH and its underlying causes from cardiovascular imaging. This systematic review aims to investigate the diagnostic performance of AI models developed to diagnose LVH and its common aetiologies.

METHODS

MEDLINE/PubMed, EMBASE and Cochrane databases were systematically searched to identify relevant studies on echocardiography, cardiac magnetic resonance (CMR), and cardiac computed tomography (CT).

RESULTS

Thirty studies were included in this review. Of them, 14 were on echocardiography, 15 on CMR, and one on cardiac CT. Regarding the AI methods applied, 79 % of studies in echocardiography utilized deep learning (DL), 64 % employed convolutional neural networks (CNNs), and 21 % applied traditional machine learning (ML) algorithms. For CMR studies, 53 % used DL, 27 % relied on CNNs, and 47 % adopted traditional ML methods. All studies showed good diagnostic performances, but those applying AI tools to determine the underlying causes of LVH demonstrated the highest accuracy metrics compared to those focused on detecting LVH itself.

CONCLUSION

AI models designed to detect and differentiate LVH on cardiac imaging are currently under development and are demonstrating promising results. Further studies focusing on real-life validation of these models, and cost-effectiveness analyses are needed.

An institutional framework to support ethical fair and equitable artificial intelligence augmented care

Coordinated access to multi-domain health data can facilitate the development and implementation of artificial intelligence-augmented clinical decision support (AI-CDS). However, scalable institutional frameworks supporting these activities are lacking. We present the PULSE framework, aimed to establish an integrative and ethically governed ecosystem for the patient-guided, patient-contextualized use of multi-domain health data for AI-augmented care. We describe deliverables related to stakeholder engagement and infrastructure development to support routine engagement of patients for consent-guided data abstraction, pre-processing, and cloud migration to support AI-CDS model development and surveillance. Central focus is placed on the routine collection of social determinants of health and patient self-reported health status to contextualize and evaluate models for fair and equitable use. Inaugural feasibility is reported for over 30,000 consecutively engaged patients. The described framework, conceptually developed to support a multi-site cardiovascular institute, is translatable to other disease domains, offering a validated architecture for use by large-scale tertiary care institutions.

A machine learning prediction model for Cardiac Amyloidosis using routine blood tests in patients with left ventricular hypertrophy

Current approaches for cardiac amyloidosis (CA) identification are time-consuming, labor-intensive, and present challenges in sensitivity and accuracy, leading to limited treatment efficacy and poor prognosis for patients. In this retrospective study, we aimed to leverage machine learning (ML) to create a diagnostic model for CA using data from routine blood tests. Our dataset included 6,563 patients with left ventricular hypertrophy, 261 of whom had been diagnosed with CA. We divided the dataset into training and testing cohorts, applying ML algorithms such as logistic regression, random forest, and XGBoost for automated learning and prediction. Our model's diagnostic accuracy was then evaluated against CA biomarkers, specifically serum-free light chains (FLCs). The model's interpretability was elucidated by visualizing the feature importance through the gain map. XGBoost outperformed both random forest and logistic regression in internal validation on the testing cohort, achieving an area under the curve (AUC) of 0.95 (95%CI: 0.92-0.97), sensitivity of 0.92 (95%CI: 0.86-0.98), specificity of 0.95 (95%CI: 0.94-0.97), and an F1 score of 0.89 (95%CI: 0.85-0.92). Its performance was also superior to the serum FLC-kappa and FLC-lambda combination (AUC of 0.88). Furthermore, XGBoost identified unique biomarker signatures indicative of multisystem dysfunction in CA patients, with significant changes in eGFR, FT3, cTnI, ANC, and NT-proBNP. This study develops a highly sensitive and accurate ML model for CA detection using routine clinical laboratory data, effectively streamlining diagnostic procedures, and providing valuable clinical insights and guiding future research into disease mechanisms.

Cardiac magnetic resonance imaging for discrimination of hypertensive heart disease and hypertrophic cardiomyopathy: a systematic review and meta-analysis

Introduction

Differentiating hypertensive heart disease (HHD) from hypertrophic cardiomyopathy (HCM) is crucial yet challenging due to overlapping clinical and morphological features. Recent studies have explored the use of various cardiac magnetic resonance (CMR) parameters to distinguish between these conditions, but findings have remained inconclusive. This study aims to identify which CMR parameters effectively discriminate between HHD and HCM and to investigate their underlying pathophysiological mechanisms through a meta-analysis.

Methods

The researchers conducted a systematic and comprehensive search for all studies that used CMR to discriminate between HHD and HCM and calculated the Hedges'g effect size for each of the included studies, which were then pooled using a random-effects model and tested for the effects of potential influencing variables through subgroup and regression analyses.

Results

In this review, 26 studies encompassing 1,349 HHD and 1,581 HCM cases were included for meta-analysis. Analysis revealed that HHD showed a significant lower in T1 mapping (g = -0.469, P < 0.001), extracellular volume (g = -0.417, P = 0.024), left ventricular mass index (g = -0.437, P < 0.001), and maximal left ventricular wall thickness (g = -2.076, P < 0.001), alongside a significant higher in end-systolic volume index (g = 0.993, P < 0.001) and end-diastolic volume index (g = 0.553, P < 0.001), compared to HCM.

Conclusion

This study clearly demonstrates that CMR parameters can effectively differentiate between HHD and HCM. HHD is characterized by significantly lower diffuse interstitial fibrosis and myocardial hypertrophy, along with better-preserved diastolic function but lower systolic function, compared to HCM. The findings highlight the need for standardized CMR protocols, considering the significant influence of MRI machine vendors, post-processing software, and study regions on diagnostic parameters. These insights are crucial for improving diagnostic accuracy and optimizing treatment strategies for patients with HHD and HCM.

Systematic Review Registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023470557, PROSPERO (CRD42023470557).

Improving the efficiency and accuracy of cardiovascular magnetic resonance with artificial intelligence-review of evidence and proposition of a roadmap to clinical translation

BACKGROUND

Cardiovascular magnetic resonance (CMR) is an important imaging modality for the assessment of heart disease; however, limitations of CMR include long exam times and high complexity compared to other cardiac imaging modalities. Recently advancements in artificial intelligence (AI) technology have shown great potential to address many CMR limitations. While the developments are remarkable, translation of AI-based methods into real-world CMR clinical practice remains at a nascent stage and much work lies ahead to realize the full potential of AI for CMR.

METHODS

Herein we review recent cutting-edge and representative examples demonstrating how AI can advance CMR in areas such as exam planning, accelerated image reconstruction, post-processing, quality control, classification and diagnosis.

RESULTS

These advances can be applied to speed up and simplify essentially every application including cine, strain, late gadolinium enhancement, parametric mapping, 3D whole heart, flow, perfusion and others. AI is a unique technology based on training models using data. Beyond reviewing the literature, this paper discusses important AI-specific issues in the context of CMR, including (1) properties and characteristics of datasets for training and validation, (2) previously published guidelines for reporting CMR AI research, (3) considerations around clinical deployment, (4) responsibilities of clinicians and the need for multi-disciplinary teams in the development and deployment of AI in CMR, (5) industry considerations, and (6) regulatory perspectives.

CONCLUSIONS

Understanding and consideration of all these factors will contribute to the effective and ethical deployment of AI to improve clinical CMR.

A Deep Learning Approach to Classify Fabry Cardiomyopathy from Hypertrophic Cardiomyopathy Using Cine Imaging on Cardiac Magnetic Resonance

A challenge in accurately identifying and classifying left ventricular hypertrophy (LVH) is distinguishing it from hypertrophic cardiomyopathy (HCM) and Fabry disease. The reliance on imaging techniques often requires the expertise of multiple specialists, including cardiologists, radiologists, and geneticists. This variability in the interpretation and classification of LVH leads to inconsistent diagnoses. LVH, HCM, and Fabry cardiomyopathy can be differentiated using T1 mapping on cardiac magnetic resonance imaging (MRI). However, differentiation between HCM and Fabry cardiomyopathy using echocardiography or MRI cine images is challenging for cardiologists. Our proposed system named the MRI short-axis view left ventricular hypertrophy classifier (MSLVHC) is a high-accuracy standardized imaging classification model developed using AI and trained on MRI short-axis (SAX) view cine images to distinguish between HCM and Fabry disease. The model achieved impressive performance, with an F1-score of 0.846, an accuracy of 0.909, and an AUC of 0.914 when tested on the Taipei Veterans General Hospital (TVGH) dataset. Additionally, a single-blinding study and external testing using data from the Taichung Veterans General Hospital (TCVGH) demonstrated the reliability and effectiveness of the model, achieving an F1-score of 0.727, an accuracy of 0.806, and an AUC of 0.918, demonstrating the model's reliability and usefulness. This AI model holds promise as a valuable tool for assisting specialists in diagnosing LVH diseases.

Current and Evolving Multimodality Cardiac Imaging in Managing Transthyretin Amyloid Cardiomyopathy

Amyloid transthyretin (ATTR) amyloidosis is a protein-misfolding disease characterized by fibril accumulation in the extracellular space that can result in local tissue disruption and organ dysfunction. Cardiac involvement drives morbidity and mortality, and the heart is the major organ affected by ATTR amyloidosis. Multimodality cardiac imaging (ie, echocardiography, scintigraphy, and cardiac magnetic resonance) allows accurate diagnosis of ATTR cardiomyopathy (ATTR-CM), and this is of particular importance because ATTR-targeting therapies have become available and probably exert their greatest benefit at earlier disease stages. Apart from establishing the diagnosis, multimodality cardiac imaging may help to better understand pathogenesis, predict prognosis, and monitor treatment response. The aim of this review is to give an update on contemporary and evolving cardiac imaging methods and their role in diagnosing and managing ATTR-CM. Further, an outlook is presented on how artificial intelligence in cardiac imaging may improve future clinical decision making and patient management in the setting of ATTR-CM.

Cardiac "hypertrophy" phenotyping: differentiating aetiologies with increased left ventricular wall thickness on echocardiography

Aims

Differentiating phenotypes of cardiac "hypertrophy" characterised by increased wall thickness on echocardiography is essential for management and prognostication. Transthoracic echocardiography is the most commonly used screening test for this purpose. We sought to identify echocardiographic markers that distinguish infiltrative and storage disorders that present with increased left ventricular (LV) wall thickness, namely, cardiac amyloidosis (CA) and Anderson-Fabry disease (AFD), from hypertensive heart disease (HHT).

Methods

Patients were retrospectively recruited from Westmead Hospital, Sydney, and Princess Alexandra Hospital, Brisbane. LV structural, systolic, and diastolic function parameters, as well as global (LVGLS) and segmental longitudinal strains, were assessed. Previously reported echocardiographic parameters including relative apical sparing ratio (RAS), LV ejection fraction-to-strain ratio (EFSR), mass-to-strain ratio (MSR) and amyloidosis index (AMYLI) score (relative wall thickness × E/e') were evaluated.

Results

A total of 209 patients {120 CA [58 transthyretin amyloidosis (ATTR) and 62 light-chain (AL) amyloidosis], 31 AFD and 58 HHT patients; mean age 64.1 ± 13.7 years, 75% male} comprised the study cohort. Echocardiographic measurements differed across the three groups, The LV mass index was higher in both CA {median 126.6 [interquartile range (IQR) 106.4-157.9 g/m²]} and AFD [median 134 (IQR 108.8-152.2 g/m²)] vs. HHT [median 92.7 (IQR 79.6-102.3 g/m²), p < 0.05]. LVGLS was lowest in CA [median 12.29 (IQR 10.33-15.56%)] followed by AFD [median 16.92 (IQR 14.14-18.78%)] then HHT [median 18.56 (IQR 17.51-19.97%), p < 0.05]. Diastolic function measurements including average e' and E/e' were most impaired in CA and least impaired in AFD. Indexed left atrial volume was highest in CA. EFSR and MSR differentiated secondary (CA + AFD) from HHT [receiver operating curve-area under the curve (ROC-AUC) of 0.80 and 0.91, respectively]. RAS and AMYLI score differentiated CA from AFD (ROC-AUC of 0.79 and 0.80, respectively). A linear discriminant analysis with stepwise variable selection using linear combinations of LV mass index, average e', LVGLS and basal strain correctly classified 79% of all cases.

Conclusion

Simple echocardiographic parameters differentiate between different "hypertrophic" cardiac phenotypes. These have potential utility as a screening tool to guide further confirmatory testing.

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Cardiac amyloidosis is an uncommon restrictive cardiomyopathy featuring an unregulated amyloid protein deposition that impairs organic function. Early cardiac amyloidosis diagnosis is generally delayed by indistinguishable clinical findings of more frequent hypertrophic diseases. Furthermore, amyloidosis is divided into various groups, according to a generally accepted taxonomy, based on the proteins that make up the amyloid deposits; a careful differentiation between the various forms of amyloidosis is necessary to undertake an adequate therapeutic treatment. Thus, cardiac amyloidosis is thought to be underdiagnosed, which delays necessary therapeutic procedures, diminishing quality of life and impairing clinical prognosis. The diagnostic work-up for cardiac amyloidosis begins with the identification of clinical features, electrocardiographic and imaging findings suggestive or compatible with cardiac amyloidosis, and often requires the histological demonstration of amyloid deposition. One approach to overcome the difficulty of an early diagnosis is the use of automated diagnostic algorithms. Machine learning enables the automatic extraction of salient information from “raw data” without the need for pre-processing methods based on the a priori knowledge of the human operator. This review attempts to assess the various diagnostic approaches and artificial intelligence computational techniques in the detection of cardiac amyloidosis.

Normative healthy reference values for global and segmental 3D principal and geometry dependent strain from cine cardiac magnetic resonance imaging

3-Dimensional (3D) myocardial deformation analysis (3D-MDA) enables novel descriptions of geometry-independent principal strain (PS). Applied to routine 2D cine cardiovascular magnetic resonance (CMR), this provides unique measures of myocardial biomechanics for disease diagnosis and prognostication. However, healthy reference values remain undefined. This study describes age- and sex-stratified reference values from CMR-based 3D-MDA, including 3D PS. One hundred healthy volunteers were prospectively recruited following institutional ethics approval and underwent CMR imaging. 3D-MDA was performed using validated software. Age- and sex-stratified global and segmental strain measures were derived for conventional geometry-dependent [circumferential (CS), longitudinal (LS), and radial (RS)] and geometry-independent [minimum (minPS) and maximum principal (maxPS)] directions of deformation. Layer-specific contraction angle interactions were determined using local minPS vectors. The average age was 43 ± 15 years and 55% were women. Strain measures were higher in women versus men. 3D PS-based assessment of maximum tissue shortening (minPS) and maximum tissue thickening (maxPS) were greater than corresponding geometry-dependent markers of LS and RS, consistent with improved representation of local tissue deformations. Global maxPS amplitude best discriminated both age and sex. Segmental analyses showed greater strain amplitudes in apical segments. Transmural PS contraction angles were higher in females and showed a heterogeneous distribution across segments. In this study we provided age and sex-based reference values for 3D strain from CMR imaging, demonstrating improved capacity for 3D PS to document maximal local tissue deformations and to discriminate age and sex phenotypes. Novel markers of layer-specific strain angles from 3D PS were also described.

Limits of Cardiovascular Adaptation During an Extreme Ultramarathon: Insights From Serial Multidimensional, Multiparametric CMR

Extreme endurance athletic challenges provide unique opportunities to study the cardiovascular system's capacity for structural, functional, and hemodynamic adaptation. The authors present a case of a male subject who ran 2,469 km, with serial multiparametric cardiac magnetic resonance imaging used to demonstrate adaptive and maladaptive alterations in cardiac remodeling and myocardial tissue health. (Level of Difficulty: Advanced.).

Association of Three-Dimensional Mesh-Derived Right Ventricular Strain with Short-Term Outcomes in Patients Undergoing Cardiac Surgery

BACKGROUND

Three-dimensional (3D) right ventricular (RV) strain analysis is not routinely performed perioperatively. Although 3D RV strain adds incrementally to outcome prediction in various cardiac diseases, its role in the perioperative setting is not sufficiently understood. The aim of this study was to investigate the association between 3D RV strain measured on RV meshes created from 3D transesophageal echocardiographic data and short-term outcomes among patients undergoing cardiac surgery.

METHODS

A total of 496 patients undergoing cardiac surgery who underwent intraoperative 3D transesophageal echocardiography (under general anesthesia, before sternotomy) were retrospectively selected, and RV meshes were generated using commercially available speckle-tracking software. Custom-made software automatically quantified longitudinal and circumferential RV strains on the mesh surfaces. Echocardiographic and clinical parameters were entered into logistic regression models to determine their associations with the primary (in-hospital death or need for extracorporeal life support) and secondary (postoperative ventilation > 48 hours) end points.

RESULTS

Mesh-derived RV strain analysis was feasible in 94% of patients and revealed distinct regional patterns with basal-apical gradients for both longitudinal and circumferential strain. Thirty-seven patients (7.6%) reached the primary end point, and 118 patients (23.8%) reached the secondary end point. In a multivariable logistic regression model, serum lactate (P < .01), an emergency indication for surgery (P < .01), tricuspid regurgitation (P < .001), and mesh-derived RV global longitudinal strain (RV-GLS; P < .01) were independently associated with the primary end point, while established measures of RV function (3D RV ejection fraction, fractional area change, tricuspid annular plane systolic excursion) and left ventricular (LV) function (3D-derived LV ejection fraction and LV-GLS) were not independently associated. Hematocrit (P < .01), serum lactate (P < .001), pulmonary hypertension (P = .04), tricuspid regurgitation (P < .01), emergency procedures (P = .02), LV-GLS (P = .02), and RV-GLS (P < .001) were associated with the secondary end point.

CONCLUSIONS

RV-GLS measured on RV meshes derived from 3D transesophageal echocardiography was independently associated with short-term outcomes in patients undergoing cardiac surgery and might be helpful for identifying patients at risk for adverse postoperative events.

Quantification of Myocardial Deformation Applying CMR-Feature-Tracking-All About the Left Ventricle?

Purpose of Review

Cardiac magnetic resonance-feature-tracking (CMR-FT)-based deformation analyses are key tools of cardiovascular imaging and applications in heart failure (HF) diagnostics are expanding. In this review, we outline the current range of application with diagnostic and prognostic implications and provide perspectives on future trends of this technique.

Recent Findings

By applying CMR-FT in different cardiovascular diseases, increasing evidence proves CMR-FT-derived parameters as powerful diagnostic and prognostic imaging biomarkers within the HF continuum partly outperforming traditional clinical values like left ventricular ejection fraction. Importantly, HF diagnostics and deformation analyses by CMR-FT are feasible far beyond sole left ventricular performance evaluation underlining the holistic nature and accuracy of this imaging approach.

Summary

As an established and continuously evolving technique with strong prognostic implications, CMR-FT deformation analyses enable comprehensive cardiac performance quantification of all cardiac chambers.

Fabry Disease and the Heart: A Comprehensive Review

Fabry disease (FD) is an X-linked lysosomal storage disorder caused by mutations of the GLA gene that result in a deficiency of the enzymatic activity of α-galactosidase A and consequent accumulation of glycosphingolipids in body fluids and lysosomes of the cells throughout the body. GB3 accumulation occurs in virtually all cardiac cells (cardiomyocytes, conduction system cells, fibroblasts, and endothelial and smooth muscle vascular cells), ultimately leading to ventricular hypertrophy and fibrosis, heart failure, valve disease, angina, dysrhythmias, cardiac conduction abnormalities, and sudden death. Despite available therapies and supportive treatment, cardiac involvement carries a major prognostic impact, representing the main cause of death in FD. In the last years, knowledge has substantially evolved on the pathophysiological mechanisms leading to cardiac damage, the natural history of cardiac manifestations, the late-onset phenotypes with predominant cardiac involvement, the early markers of cardiac damage, the role of multimodality cardiac imaging on the diagnosis, management and follow-up of Fabry patients, and the cardiac efficacy of available therapies. Herein, we provide a comprehensive and integrated review on the cardiac involvement of FD, at the pathophysiological, anatomopathological, laboratory, imaging, and clinical levels, as well as on the diagnosis and management of cardiac manifestations, their supportive treatment, and the cardiac efficacy of specific therapies, such as enzyme replacement therapy and migalastat.