Deep Learning-Based Prediction of Right Ventricular Ejection Fraction Using 2D Echocardiograms

Assessing Right Ventricular Function in the Perioperative Setting, Part I: Echo-Based Measurements

This article explores the growing significance of right ventricular (RV) function, particularly in perioperative settings. The right ventricle plays a crucial role in predicting morbidity and mortality, especially in cardiac surgeries. Right ventricular failure is associated with high in-hospital mortality, making accurate assessment vital. The article discusses echocardiographic evaluation, emphasizing both qualitative and quantitative measures, including tricuspid annular plane systolic excursion, fractional area change, and myocardial strain imaging. Understanding RV pathophysiology is essential for effective diagnosis and management, particularly in dynamic perioperative conditions influenced by ventilation, anesthesia, and extracorporeal circulation.

Machine learning prediction of right ventricular volume and ejection fraction from two-dimensional echocardiography in patients with pulmonary regurgitation

Right ventricular (RV) end-diastolic volume (RVEDV) and ejection fraction (RVEF) by cardiac MRI (cMRI) guide management in chronic pulmonary regurgitation (PR). Two-dimensional echocardiography suboptimally correlate with RV volumes. This study tested whether combination of guideline-directed RV measures in a machine learning (ML) framework improves quantitative assessment of RVEDV and RVEF. RV measurements were obtained on subjects with > mild PR who had cMRI and echocardiogram within 90 days. A gradient-boosted trees algorithm predicted cMRI RV dilation (RVEDV > 160 ml/m2) and RV dysfunction (RVEF<47%), first with "guideline-only" measures, and then with "expanded-features" to include 44 total echocardiographic, clinical, and demographic variables. Model performance was compared to clinician visual assessment. Of 232 studies (56% tetralogy of Fallot, 20% pulmonary stenosis), the median age was 21.5 years, 21 (9%) had RV dilation, and 42 (18%) had RV dysfunction. For RV dilation prediction, the guideline-only model area under the receiver operating characteristic (AUROC)=0.68, and expanded-features model AUROC=0.85. At 90% sensitivity, the expanded-features model had 73% specificity, 25% positive predictive value (PPV), and 99% negative predictive value (NPV) This was similar to clinician performance (sensitivity 81%, specificity 81%, PPV 29%, NPV 98%). For prediction of RV dysfunction, the guideline-only AUROC= 0.71, additional features did not improve the model, and clinicians outperformed the model. In patients with PR, a ML model combining guidelines for RV assessment with demographic and additional echocardiographic parameters may effectively rule-out those with significant RV dilation at clinical thresholds for intervention, and performs similarly to expert clinicians.

Hearts, Data, and Artificial Intelligence Wizardry: From Imitation to Innovation in Cardiovascular Care

Artificial intelligence (AI) is transforming cardiovascular medicine by enabling the analysis of high-dimensional biomedical data with unprecedented precision. Initially employed to automate human tasks such as electrocardiogram (ECG) interpretation and imaging segmentation, AI's true potential lies in uncovering hidden disease data patterns, predicting long-term cardiovascular risk, and personalizing treatments. Unlike human cognition, which excels in certain tasks but is limited by memory and processing constraints, AI integrates multimodal data sources-including ECG, echocardiography, cardiac magnetic resonance (CMR) imaging, genomics, and wearable sensor data-to generate novel clinical insights. AI models have demonstrated remarkable success in early dis-ease detection, such as predicting heart failure from standard ECGs before symptom on-set, distinguishing genetic cardiomyopathies, and forecasting arrhythmic events. However, several challenges persist, including AI's lack of contextual understanding in most of these tasks, its "black-box" nature, and biases in training datasets that may contribute to disparities in healthcare delivery. Ethical considerations and regulatory frameworks are evolving, with governing bodies establishing guidelines for AI-driven medical applications. To fully harness the potential of AI, interdisciplinary collaboration among clinicians, data scientists, and engineers is essential, alongside open science initiatives to promote data accessibility and reproducibility. Future AI models must go beyond task automation, focusing instead on augmenting human expertise to enable proactive, precision-driven cardiovascular care. By embracing AI's computational strengths while addressing its limitations, cardiology is poised to enter an era of transformative innovation beyond traditional diagnostic and therapeutic paradigms.

PanEcho: Complete AI-enabled echocardiography interpretation with multi-task deep learning

Importance

Echocardiography is a cornerstone of cardiovascular care but relies on expert interpretation and manual reporting from a series of videos. We propose an artificial intelligence (AI) system, PanEcho, to automate echocardiogram interpretation with multi-task deep learning.

Objective

To develop and evaluate the accuracy of PanEcho on a comprehensive set of 39 echocardiographic labels and measurements on transthoracic echocardiography (TTE).

Design Setting and Participants

This study represents the development and retrospective, multi-site validation of an AI system. PanEcho was developed using a sample of TTE studies conducted at Yale-New Haven Health System (YNHHS) hospitals and clinics from January 2016-June 2022 during routine care. The trained model was internally validated in a temporally distinct YNHHS cohort from July-December 2022, externally validated across four diverse external cohorts, and made publicly available.

Main Outcomes and Measures

The primary outcome was the area under the receiver operating characteristic curve (AUC) for diagnostic classification tasks and mean absolute error (MAE) for parameter estimation tasks, comparing AI predictions with the assessment of the interpreting cardiologist.

Results

This study included 1.2 million echocardiographic videos from 32,265 TTE studies of 24,405 patients across YNHHS hospitals and clinics. PanEcho performed 18 diagnostic classification tasks with a median AUC of 0.91 (IQR: 0.88-0.93) and estimated 21 echocardiographic parameters with a median normalized MAE of 0.13 (0.10-0.18) in internal validation. For instance, the model accurately estimated left ventricular (LV) ejection fraction (MAE: 4.2% internal; 4.5% external) and detected moderate or higher LV systolic dysfunction (AUC: 0.98 internal; 0.99 external), RV systolic dysfunction (0.93 internal; 0.94 external), and severe aortic stenosis (0.98 internal; 1.00 external). PanEcho maintained excellent performance in limited imaging protocols, performing 15 diagnosis tasks with 0.91 median AUC (IQR: 0.87-0.94) in an abbreviated TTE cohort and 14 tasks with 0.85 median AUC (0.77-0.87) on real-world point-of-care ultrasound acquisitions by non-experts from YNHHS emergency departments.

Conclusions and Relevance

We report an AI system that automatically interprets echocardiograms, maintaining high accuracy across geography and time from complete and limited studies. PanEcho may be used as an adjunct reader in echocardiography labs or rapid AI-enabled screening tool in point-of-care settings.

Artificial Intelligence in Diagnosis of Heart Failure

Heart failure (HF) is a complex and varied condition that affects over 50 million people worldwide. Although there have been significant strides in understanding the underlying mechanisms of HF, several challenges persist, particularly in the accurate diagnosis of HF. These challenges include issues related to its classification, the identification of specific phenotypes, and the assessment of disease severity. Artificial intelligence (AI) algorithms have the potential to transform HF care by enhancing clinical decision-making processes, enabling the early detection of patients at risk for subclinical or worsening HF. By integrating and analyzing vast amounts of data with intricate multidimensional interactions, AI algorithms can provide critical insights that help physicians make more timely and informed decisions. In this review, we explore the challenges in current diagnosis of HF, basic AI concepts and common AI algorithms, and latest AI research in HF diagnosis.

Deep Learning-Enabled Assessment of Right Ventricular Function Improves Prognostication After Transcatheter Edge-to-Edge Repair for Mitral Regurgitation

BACKGROUND

Right ventricular (RV) function has a well-established prognostic role in patients with severe mitral regurgitation (MR) undergoing transcatheter edge-to-edge repair (TEER) and is typically assessed using echocardiography-measured tricuspid annular plane systolic excursion. Recently, a deep learning model has been proposed that accurately predicts RV ejection fraction (RVEF) from 2-dimensional echocardiographic videos, with similar diagnostic accuracy as 3-dimensional imaging. This study aimed to evaluate the prognostic value of the deep learning-predicted RVEF values in patients with severe MR undergoing TEER.

METHODS

This multicenter registry study analyzed the associations between the predicted RVEF values and 1-year mortality in patients with severe MR undergoing TEER. To predict RVEF, 2-dimensional apical 4-chamber view videos from preprocedural transthoracic echocardiographic studies were exported and processed by a rigorously validated deep learning model.

RESULTS

Good-quality 2-dimensional apical 4-chamber view videos could be retrieved for 1154 patients undergoing TEER between 2017 and 2023. Survival at 1 year after TEER was 84.7%. The predicted RVEF values ranged from 26.6% to 64.0% and correlated only modestly with tricuspid annular plane systolic excursion (Pearson R=0.33; P<0.001). Importantly, predicted RVEF was superior to tricuspid annular plane systolic excursion levels in predicting 1-year mortality after TEER (area under the curve, 0.687 versus 0.625; P=0.029). Furthermore, Kaplan-Meier survival analysis revealed that patients with reduced RV function (n=723; defined as a predicted RVEF of <45%) had significantly worse 1-year survival rates than patients with preserved RV function (n=431; defined as a predicted RVEF of ≥45%; 80.3% [95% CI, 77.4%-83.3%] versus 92.1% [95% CI, 89.5%-94.7%]; hazard ratio for 1-year mortality, 2.67 [95% CI, 1.82-3.90]; P<0.001).

CONCLUSIONS

Deep learning-enabled assessment of RV function using standard 2-dimensional echocardiographic videos can refine the prognostication of patients with severe MR undergoing TEER. Thus, it can be used to screen for patients with RV dysfunction who might benefit from intensified follow-up care.

Emerging multimodality imaging techniques for the pulmonary circulation

Pulmonary hypertension (PH) remains a challenging condition to diagnose, classify and treat. Current approaches to the assessment of PH include echocardiography, ventilation/perfusion scintigraphy, cross-sectional imaging using computed tomography and magnetic resonance imaging, and right heart catheterisation. However, these approaches only provide an indirect readout of the primary pathology of the disease: abnormal vascular remodelling in the pulmonary circulation. With the advent of newer imaging techniques, there is a shift toward increased utilisation of noninvasive high-resolution modalities that offer a more comprehensive cardiopulmonary assessment and improved visualisation of the different components of the pulmonary circulation. In this review, we explore advances in imaging of the pulmonary vasculature and their potential clinical translation. These include advances in diagnosis and assessing treatment response, as well as strategies that allow reduced radiation exposure and implementation of artificial intelligence technology. These emerging modalities hold the promise of developing a deeper understanding of pulmonary vascular disease and the impact of comorbidities. They also have the potential to improve patient outcomes by reducing time to diagnosis, refining classification, monitoring treatment response and improving our understanding of disease mechanisms.

Accurate fully automated assessment of left ventricle, left atrium, and left atrial appendage function from computed tomography using deep learning

Aims

Assessment of cardiac function is essential for diagnosis and treatment planning in cardiovascular disease. Volume of cardiac regions and the derived measures of stroke volume (SV) and ejection fraction (EF) are most accurately calculated from imaging. This study aims to develop a fully automatic deep learning approach for calculation of cardiac function from computed tomography (CT).

Methods and results

Time-resolved CT data sets from 39 patients were used to train segmentation models for the left side of the heart including the left ventricle (LV), left atrium (LA), and left atrial appendage (LAA). We compared nnU-Net, 3D TransUNet, and UNETR. Dice Similarity Scores (DSS) were similar between nnU-Net (average DSS = 0.91) and 3D TransUNet (DSS = 0.89) while UNETR performed less well (DSS = 0.69). Intra-class correlation analysis showed nnU-Net and 3D TransUNet both accurately estimated LVSV (ICCnnU-Net = 0.95; ICC3DTransUNet = 0.94), LVEF (ICCnnU-Net = 1.00; ICC3DTransUNet = 1.00), LASV (ICCnnU-Net = 0.91; ICC3DTransUNet = 0.80), LAEF (ICCnnU-Net = 0.95; ICC3DTransUNet = 0.81), and LAASV (ICCnnU-Net = 0.79; ICC3DTransUNet = 0.81). Only nnU-Net significantly predicted LAAEF (ICCnnU-Net = 0.68). UNETR was not able to accurately estimate cardiac function. Time to convergence during training and time needed for inference were both faster for 3D TransUNet than for nnU-Net.

Conclusion

nnU-Net outperformed two different vision transformer architectures for the segmentation and calculation of function parameters for the LV, LA, and LAA. Fully automatic calculation of cardiac function parameters from CT using deep learning is fast and reliable.

Significant Disagreement Between Conventional Parameters and 3D Echocardiography-Derived Ejection Fraction in the Detection of Right Ventricular Systolic Dysfunction and Its Association With Outcomes

AIMS

Conventional echocardiographic parameters such as tricuspid annular plane systolic excursion (TAPSE), fractional area change (FAC), and free-wall longitudinal strain (FWLS) offer limited insights into the complexity of right ventricular (RV) systolic function, while 3D echocardiography-derived RV ejection fraction (RVEF) enables a comprehensive assessment. We investigated the discordance between TAPSE, FAC, FWLS, and RVEF in RV systolic function grading and associated outcomes.

METHODS

We analyzed two- and three-dimensional echocardiography data from 2 centers including 750 patients followed up for all-cause mortality. Right ventricular dysfunction was defined as RVEF <45%, with guideline-recommended thresholds (TAPSE <17 mm, FAC <35%, FWLS >-20%) considered.

RESULTS

Among patients with normal RVEF, significant proportions exhibited impaired TAPSE (21%), FAC (33%), or FWLS (8%). Conversely, numerous patients with reduced RVEF had normal TAPSE (46%), FAC (26%), or FWLS (41%). Using receiver-operating characteristic analysis, FWLS exhibited the highest area under the curve of discrimination for RV dysfunction (RVEF <45%) with 59% sensitivity and 92% specificity. Over a median 3.7-year follow-up, 15% of patients died. Univariable Cox regression identified TAPSE, FAC, FWLS, and RVEF as significant mortality predictors. Combining impaired conventional parameters showed that outcomes are the worst if at least 2 parameters are impaired and gradually better if only one or none of them are impaired (log-rank P < .005).

CONCLUSION

Guideline-recommended cutoff values of conventional echocardiographic parameters of RV systolic function are only modestly associated with RVEF-based assessment. Impaired values of FWLS showed the closest association with the RVEF cutoff. Our results emphasize a multiparametric approach in the assessment of RV function, especially if 3D echocardiography is not available.

Deep learning to assess right ventricular ejection fraction from two-dimensional echocardiograms in precapillary pulmonary hypertension

BACKGROUND

Precapillary pulmonary hypertension (PH) is characterized by a sustained increase in right ventricular (RV) afterload, impairing systolic function. Two-dimensional (2D) echocardiography is the most performed cardiac imaging tool to assess RV systolic function; however, an accurate evaluation requires expertise. We aimed to develop a fully automated deep learning (DL)-based tool to estimate the RV ejection fraction (RVEF) from 2D echocardiographic videos of apical four-chamber views in patients with precapillary PH.

METHODS

We identified 85 patients with suspected precapillary PH who underwent cardiac magnetic resonance imaging (MRI) and echocardiography. The data was divided into training (80%) and testing (20%) datasets, and a regression model was constructed using 3D-ResNet50. Accuracy was assessed using five-fold cross validation.

RESULTS

The DL model predicted the cardiac MRI-derived RVEF with a mean absolute error of 7.67%. The DL model identified severe RV systolic dysfunction (defined as cardiac MRI-derived RVEF < 37%) with an area under the curve (AUC) of .84, which was comparable to the AUC of RV fractional area change (FAC) and tricuspid annular plane systolic excursion (TAPSE) measured by experienced sonographers (.87 and .72, respectively). To detect mild RV systolic dysfunction (defined as RVEF ≤ 45%), the AUC from the DL-predicted RVEF also demonstrated a high discriminatory power of .87, comparable to that of FAC (.90), and significantly higher than that of TAPSE (.67).

CONCLUSION

The fully automated DL-based tool using 2D echocardiography could accurately estimate RVEF and exhibited a diagnostic performance for RV systolic dysfunction comparable to that of human readers.

Effect of left bundle branch pacing on right ventricular function: A 3-dimensional echocardiography study

BACKGROUND

The effect of left bundle branch pacing (LBBP) on right ventricular (RV) function is not well known, and there is conflicting evidence regarding whether cardiac resynchronization therapy improves RV function.

OBJECTIVES

The study aimed to investigate the effect of LBBP on RV function and to evaluate the response of RV dysfunction (RVD) to LBBP.

METHODS

Sixty-five LBBP candidates were prospectively included in the study and underwent echocardiography at baseline and 6-month follow-up. LBBP response was left ventricular (LV) reverse remodeling, defined as a reduction in LV end-systolic volume of ≥15% at follow-up.

RESULTS

Patients were assigned to 2 subgroups on the basis of 3-dimensional echocardiography-derived RV ejection fraction (EF) before LBBP implantation: 30 patients (46%) in the no RVD group and 35 patients (54%) in the RVD group. The RVD group was characterized by higher N-terminal pro-brain natriuretic peptide levels, New York Heart Association functional class, and larger LV/RV size. LBBP induced a significant reduction in QRS duration, LV size, and improvement in LVEF and mechanical dyssynchrony in both the no RVD and RVD groups, and a significant improvement in RV volumes and RVEF in the RVD group (all P<.01). LBBP resulted in a similar percentage reduction in QRS duration, LV dimensions, LV volumes, and percentage improvement in LVEF in RVD and no RVD groups (all P＞.05). LV reverse remodeling (29 of 35 patients vs 27 of 30 patients; P = .323) in the RVD group was similar to that in the no RVD group after LBBP.

CONCLUSION

LBBP induces excellent electrical and mechanical resynchronization, with a significant improvement in RV volumes and function. RVD did not diminish the beneficial effects on LV reverse remodeling after LBBP.

An attention-based deep learning method for right ventricular quantification using 2D echocardiography: Feasibility and accuracy

AIM

To test the feasibility and accuracy of a new attention-based deep learning (DL) method for right ventricular (RV) quantification using 2D echocardiography (2DE) with cardiac magnetic resonance imaging (CMR) as reference.

METHODS AND RESULTS

We retrospectively analyzed images from 50 adult patients (median age 51, interquartile range 32-62 42% women) who had undergone CMR within 1 month of 2DE. RV planimetry of the myocardial border was performed in end-diastole (ED) and end-systole (ES) for eight standardized 2DE RV views with calculation of areas. The DL model comprised a Feature Tokenizer module and a stack of Transformer layers. Age, gender and calculated areas were used as inputs, and the output was RV volume in ED/ES. The dataset was randomly split into training, validation and testing subsets (35, 5 and 10 patients respectively). Mean RVEDV, RVESV and RV ejection fraction (EF) were 163 ± 70 mL, 82 ± 42 mL and 51% ± 8% respectively without differences among the subsets. The proposed method achieved good prediction of RV volumes (R2  = .953, absolute percentage error [APE] = 9.75% ± 6.23%) and RVEF (APE = 7.24% ± 4.55%). Per CMR, there was one patient with RV dilatation and three with RV dysfunction in the testing dataset. The DL model detected RV dilatation in 1/1 case and RV dysfunction in 4/3 cases.

CONCLUSIONS

An attention-based DL method for 2DE RV quantification showed feasibility and promising accuracy. The method requires validation in larger cohorts with wider range of RV size and function. Further research will focus on the reduction of the number of required 2DE to make the method clinically applicable.

Added value of 3D echocardiography in the diagnosis and prognostication of patients with right ventricular dysfunction

Recent inroads into percutaneous-based options for the treatment of tricuspid valve disease has brought to light how little we know about the behavior of the right ventricle in both health and disease and how incomplete our assessment of right ventricular (RV) physiology and function is using current non-invasive technology, in particular echocardiography. The purpose of this review is to provide an overview of what three-dimensional echocardiography (3DE) can offer currently to enhance RV evaluation and what the future may hold if we continue to improve the 3D evaluation of the right heart.

The Echocardiographic Evaluation of the Right Heart: Current and Future Advances

PURPOSE OF REVIEW

To discuss physiologic and methodologic advances in the echocardiographic assessment of right heart (RH) function, including the emergence of artificial intelligence (AI) and point-of-care ultrasound.

RECENT FINDINGS

Recent studies have highlighted the prognostic value of right ventricular (RV) longitudinal strain, RV end-systolic dimensions, and right atrial (RA) size and function in pulmonary hypertension and heart failure. While RA pressure is a central marker of right heart diastolic function, the recent emphasis on venous excess imaging (VExUS) has provided granularity to the systemic consequences of RH failure. Several methodological advances are also changing the landscape of RH imaging including post-processing 3D software to delineate the non-longitudinal (radial, anteroposterior, and circumferential) components of RV function, as well as AI segmentation- and non-segmentation-based quantification. Together with recent guidelines and advances in AI technology, the field is shifting from specific RV functional metrics to integrated RH disease-specific phenotypes. A modern echocardiographic evaluation of RH function should focus on the entire cardiopulmonary venous unit-from the venous to the pulmonary arterial system. Together, a multi-parametric approach, guided by physiology and AI algorithms, will help define novel integrated RH profiles for improved disease detection and monitoring.