Artificial intelligence-based classification of echocardiographic views.

Aims: Augmenting echocardiography with artificial intelligence would allow for automated assessment of routine parameters and identification of disease patterns not easily recognized otherwise. View classification is an essential first step before deep learning can be applied to the echocardiogram. Methods and results: We trained two- and three-dimensional convolutional neural networks (CNNs) using transthoracic echocardiographic (TTE) studies obtained from 909 patients to classify nine view categories (10 269 videos). Transthoracic echocardiographic studies from 229 patients were used in internal validation (2582 videos). Convolutional neural networks were tested on 100 patients with comprehensive TTE studies (where the two examples chosen by CNNs as most likely to represent a view were evaluated) and 408 patients with five view categories obtained via point-of-care ultrasound (POCUS). The overall accuracy of the two-dimensional CNN was 96.8%, and the averaged area under the curve (AUC) was 0.997 on the comprehensive TTE testing set; these numbers were 98.4% and 0.998, respectively, on the POCUS set. For the three-dimensional CNN, the accuracy and AUC were 96.3% and 0.998 for full TTE studies and 95.0% and 0.996 on POCUS videos, respectively. The positive predictive value, which defined correctly identified predicted views, was higher with two-dimensional rather than three-dimensional networks, exceeding 93% in apical, short-axis aortic valve, and parasternal long-axis left ventricle views. Conclusion: An automated view classifier utilizing CNNs was able to classify cardiac views obtained using TTE and POCUS with high accuracy. The view classifier will facilitate the application of deep learning to echocardiography.

Machine Learning for Diagnosis of Pulmonary Hypertension by Echocardiography.

OBJECTIVE: To evaluate a machine learning (ML)-based model for pulmonary hypertension (PH) prediction using measurements and impressions made during echocardiography. METHODS: A total of 7853 consecutive patients with right-sided heart catheterization and transthoracic echocardiography performed within 1 week from January 1, 2012, through December 31, 2019, were included. The data were split into training (n=5024 [64%]), validation (n=1275 [16%]), and testing (n=1554 [20%]). A gradient boosting machine with enumerated grid search for optimization was selected to allow missing data in the boosted trees without imputation. The training target was PH, defined by right-sided heart catheterization as mean pulmonary artery pressure above 20 mm Hg; model performance was maximized relative to area under the receiver operating characteristic curve using 5-fold cross-validation. RESULTS: Cohort age was 64±14 years; 3467 (44%) were female, and 81% (6323/7853) had PH. The final trained model included 19 characteristics, measurements, or impressions derived from the echocardiogram. In the testing data, the model had high discrimination for the detection of PH (area under the receiver operating characteristic curve, 0.83; 95% CI, 0.80 to 0.85). The model's accuracy, sensitivity, positive predictive value, and negative predictive value were 82% (1267/1554), 88% (1098/1242), 89% (1098/1241), and 54% (169/313), respectively. CONCLUSION: By use of ML, PH could be predicted on the basis of clinical and echocardiographic variables, without tricuspid regurgitation velocity. Machine learning methods appear promising for identifying patients with low likelihood of PH.

Using PAH-SYMPACT to assess quality of life in patients with pulmonary hypertension associated with chronic lung disease.

Chronic lung disease (CLD) is the second leading cause of pulmonary hypertension (PH) and is associated with significant morbidity and mortality. Although PH associated with CLD (PH-CLD) leads to impaired health-related quality of life (HRQOL), there are no validated tools to assess HRQOL in PH-CLD. The Pulmonary Arterial Hypertension-Symptoms and Impact Questionnaire (PAH-SYMPACT) is an HRQOL instrument aimed at assessing the symptoms and impact of PH on overall function and well-being. We performed a single-center prospective cohort study using PAH-SYMPACT scores to compare symptoms, exercise capacity and HRQOL in patients with PAH and PH-CLD. One hundred and twenty-five patients (99 patients with idiopathic/heritable PAH and 26 with PH-CLD) completed the PAH-SYMPACT questionnaire which consists of 22 questions that assess HRQOL across four domains: cardiopulmonary (CP) symptoms, cardiovascular (CV) symptoms, physical impact (PI), and cognitive/emotional (CE) impact. Higher scores indicate worse HRQOL. We compared patients with PAH and PH-CLD using a Wilcoxon rank sum or chi-squared test as appropriate. Multivariate linear regression analysis was used to assess the relationship between PH classification and SYMPACT scores. Compared to PAH, patients with PH-CLD were older, more likely to use oxygen and had worse functional class and exercise capacity. While there was no significant difference between the two groups in CP, CV, or CE domain scores, patients with PH-CLD had significantly worse PI scores by univariate (1.79 vs. 1.13, p < 0.001) and multivariate analysis (1.61 vs. 1.17, p = 0.02) and overall worse SYMPACT scores (1.19 vs. 0.91, p = 0.03). In conclusion, patients with PH-CLD have worse HRQOL as assessed by the PAH-SYMPACT questionnaire versus patients with PAH. Although PAH-SYMPACT has not been validated in PH-CLD, the results of this study can guide clinicians in understanding the symptoms and impact of PH-CLD relative to PAH.

Artificial intelligence-enabled ECG for left ventricular diastolic function and filling pressure.

Assessment of left ventricular diastolic function plays a major role in the diagnosis and prognosis of cardiac diseases, including heart failure with preserved ejection fraction. We aimed to develop an artificial intelligence (AI)-enabled electrocardiogram (ECG) model to identify echocardiographically determined diastolic dysfunction and increased filling pressure. We trained, validated, and tested an AI-enabled ECG in 98,736, 21,963, and 98,763 patients, respectively, who had an ECG and echocardiographic diastolic function assessment within 14 days with no exclusion criteria. It was also tested in 55,248 patients with indeterminate diastolic function by echocardiography. The model was evaluated using the area under the curve (AUC) of the receiver operating characteristic curve, and its prognostic performance was compared to echocardiography. The AUC for detecting increased filling pressure was 0.911. The AUCs to identify diastolic dysfunction grades ≥1, ≥2, and 3 were 0.847, 0.911, and 0.943, respectively. During a median follow-up of 5.9 years, 20,223 (20.5%) died. Patients with increased filling pressure predicted by AI-ECG had higher mortality than those with normal filling pressure, after adjusting for age, sex, and comorbidities in the test group (hazard ratio (HR) 1.7, 95% CI 1.645-1.757) similar to echocardiography and in the indeterminate group (HR 1.34, 95% CI 1.298-1.383). An AI-enabled ECG identifies increased filling pressure and diastolic function grades with a good prognostic value similar to echocardiography. AI-ECG is a simple and promising tool to enhance the detection of diseases associated with diastolic dysfunction and increased diastolic filling pressure.

Incidence of Severe Adverse Drug Reactions to Ultrasound Enhancement Agents in a Contemporary Echocardiography Practice.

OBJECTIVES: Prior data indicate a very rare risk of serious adverse drug reaction (ADR) to ultrasound enhancement agents (UEAs). We sought to evaluate the frequency of ADR to UEA administration in contemporary practice. METHODS: We retrospectively reviewed 4 US health systems to characterize the frequency and severity of ADR to UEA. Adverse drug reactions were considered severe when cardiopulmonary involvement was present and critical when there was loss of consciousness, loss of pulse, or ST-segment elevation. Rates of isolated back pain and headache were derived from the Mayo Clinic Rochester stress echocardiography database where systematic prospective reporting of ADR was performed. RESULTS: Among 26,539 Definity and 11,579 Lumason administrations in the Mayo Clinic Rochester stress echocardiography database, isolated back pain or headache was more frequent with Definity (0.49% vs 0.04%, P < .0001) but less common with Definity infusion versus bolus (0.08% vs 0.53%, P = .007). Among all sites there were 201,834 Definity and 84,943 Lumason administrations. Severe and critical ADR were more frequent with Lumason than with Definity (0.0848% vs 0.0114% and 0.0330% vs 0.0010%, respectively; P < .001 for each). Among the 3 health systems with >2,000 Lumason administrations, the frequency of severe ADR with Lumason ranged from 0.0755% to 0.1093% and the frequency of critical ADR ranged from 0.0293% to 0.0525%. Severe ADR rates with Definity were stable over time but increased in more recent years with Lumason (P = .02). Patients with an ADR to Lumason since the beginning of 2021 were more likely to have received a COVID-19 vaccination compared with matched controls (88% vs 75%; P = .05) and more likely to have received Moderna than Pfizer-Biotech (71% vs 26%, P < .001). CONCLUSION: Severe and critical ADR, while rare, were more frequent with Lumason, and the frequency has increased in more recent years. Additional work is needed to better understand factors, including associations with recently developed mRNA vaccines, which may be contributing to the increased rates of ADR to UEA since 2021.

Echocardiography-Based Deep Learning Model to Differentiate Constrictive Pericarditis and Restrictive Cardiomyopathy.

BACKGROUND: Constrictive pericarditis (CP) is an uncommon but reversible cause of diastolic heart failure if appropriately identified and treated. However, its diagnosis remains a challenge for clinicians. Artificial intelligence may enhance the identification of CP. OBJECTIVES: The authors proposed a deep learning approach based on transthoracic echocardiography to differentiate CP from restrictive cardiomyopathy. METHODS: Patients with a confirmed diagnosis of CP and cardiac amyloidosis (CA) (as the representative disease of restrictive cardiomyopathy) at Mayo Clinic Rochester from January 2003 to December 2021 were identified to extract baseline demographics. The apical 4-chamber view from transthoracic echocardiography studies was used as input data. The patients were split into a 60:20:20 ratio for training, validation, and held-out test sets of the ResNet50 deep learning model. The model performance (differentiating CP and CA) was evaluated in the test set with the area under the curve. GradCAM was used for model interpretation. RESULTS: A total of 381 patients were identified, including 184 (48.3%) CP, and 197 (51.7%) CA cases. The mean age was 68.7 ± 11.4 years, and 72.8% were male. ResNet50 had a performance with an area under the curve of 0.97 to differentiate the 2-class classification task (CP vs CA). The GradCAM heatmap showed activation around the ventricular septal area. CONCLUSIONS: With a standard apical 4-chamber view, our artificial intelligence model provides a platform to facilitate the detection of CP, allowing for improved workflow efficiency and prompt referral for more advanced evaluation and intervention of CP.

Prognostic Significance of Elevated Left Ventricular Filling Pressures with Exercise: Insights from a Cohort of 14,338 Patients.

BACKGROUND: Exercise echocardiography can assess for cardiovascular causes of dyspnea other than coronary artery disease. However, the prevalence and prognostic significance of elevated left ventricular (LV) filling pressures with exercise is understudied. METHODS: We evaluated 14,338 patients referred for maximal symptom-limited treadmill echocardiography. In addition to assessment of LV regional wall motion abnormalities (RWMAs), we measured patients' early diastolic mitral inflow (E), septal mitral annulus relaxation (e'), and peak tricuspid regurgitation velocity before and immediately after exercise. RESULTS: Over a mean follow-up of 3.3 ± 3.4 years, patients with E/e' ≥15 with exercise (n = 1,323; 9.2%) had lower exercise capacity (7.3 ± 2.1 vs 9.1 ± 2.4 metabolic equivalents, P < .0001) and were more likely to have resting or inducible RWMAs (38% vs 18%, P < .0001). Approximately 6% (n = 837) had elevated LV filling pressures without RWMAs. Patients with a poststress E/e' ≥15 had a 2.71-fold increased mortality rate (2.28-3.21, P < .0001) compared with those with poststress E/e' ≤ 8. Those with an E/e' of 9 to 14, while at lower risk than the E/e' ≥15 cohort (hazard ratio [HR] = 0.58 [0.48-0.69]; P < .0001), had higher risk than if E/e' ≤8 (HR = 1.56 [1.37-1.78], P < .0001). On multivariable analysis, adjusting for age, sex, exercise capacity, LV ejection fraction, and presence of pulmonary hypertension with stress, patients with E/e' ≥15 had a 1.39-fold (95% CI, 1.18-1.65, P < .0001) increased risk of all-cause mortality compared with patients without elevated LV filling pressures. Compared with patients with E/e' ≤ 15 after exercise, patients with E/e' ≤15 at rest but elevated after exercise had a higher risk of cardiovascular death (HR = 8.99 [4.7-17.3], P < .0001). CONCLUSION: Patients with elevated LV filling pressures are at increased risk of death, irrespective of myocardial ischemia or LV systolic dysfunction. These findings support the routine incorporation of LV filling pressure assessment, both before and immediately following stress, into the evaluation of patients referred for exercise echocardiography.

Challenges and solutions of echocardiography generalization for deep learning: a study in patients with constrictive pericarditis.

Purpose: The inherent characteristics of transthoracic echocardiography (TTE) images such as low signal-to-noise ratio and acquisition variations can limit the direct use of TTE images in the development and generalization of deep learning models. As such, we propose an innovative automated framework to address the common challenges in the process of echocardiography deep learning model generalization on the challenging task of constrictive pericarditis (CP) and cardiac amyloidosis (CA) differentiation. Approach: Patients with a confirmed diagnosis of CP or CA and normal cases from Mayo Clinic Rochester and Arizona were identified to extract baseline demographics and the apical 4 chamber view from TTE studies. We proposed an innovative preprocessing and image generalization framework to process the images for training the ResNet50, ResNeXt101, and EfficientNetB2 models. Ablation studies were conducted to justify the effect of each proposed processing step in the final classification performance. Results: The models were initially trained and validated on 720 unique TTE studies from Mayo Rochester and further validated on 225 studies from Mayo Arizona. With our proposed generalization framework, EfficientNetB2 generalized the best with an average area under the curve (AUC) of 0.96 (±0.01) and 0.83 (±0.03) on the Rochester and Arizona test sets, respectively. Conclusions: Leveraging the proposed generalization techniques, we successfully developed an echocardiography-based deep learning model that can accurately differentiate CP from CA and normal cases and applied the model to images from two sites. The proposed framework can be further extended for the development of echocardiography-based deep learning models.

Stress Echo 2030: the new ABCDE protocol defining the future of cardiac imaging.

Functional testing with stress echocardiography is based on the detection of regional wall motion abnormality with two-dimensional echocardiography and is embedded in clinical guidelines. Yet, it under-uses the unique versatility of the technique, ideally suited to describe the different functional abnormalities underlying the same wall motion response during stress. Five parameters converge conceptually and methodologically in the state-of-the-art ABCDE protocol, assessing multiple vulnerabilities of the ischemic patient. The five steps of the ABCDE protocol are (1) step A: regional wall motion; (2) step B: B-lines by lung ultrasound assessing extravascular lung water; (3) step C: left ventricular contractile reserve by volumetric two-dimensional echocardiography; (4) step D: coronary flow velocity reserve in mid-distal left anterior descending coronary with pulsed-wave Doppler; and (5) step E: assessment of heart rate reserve with a one-lead electrocardiogram. ABCDE stress echo offers insight into five functional reserves: epicardial flow (A); diastolic (B), contractile (C), coronary microcirculatory (D), and chronotropic reserve (E). The new format is more comprehensive and allows better functional characterization, risk stratification, and personalized tailoring of therapy. ABCDE protocol is an 'ecumenic' and 'omnivorous' functional test, suitable for all stresses and all patients also beyond coronary artery disease. It fits the need for sustainability of the current era in healthcare, since it requires universally available technology, and is low-cost, radiation-free, and nearly carbon-neutral.

The Evolving Etiologic and Epidemiologic Portrait of Pericardial Disease.

Pericardial disease includes a variety of conditions, including inflammatory pericarditis, pericardial effusions, constrictive pericarditis, pericardial cysts, and primary and secondary pericardial neoplasms. The true incidence of this varied condition is not well established, and the causes vary greatly across the world. This review aims to describe the changing pattern of epidemiology of pericardial disease and to provide an overview of causative etiologies. Idiopathic pericarditis (assumed most often to be viral) remains the most common etiology for pericardial disease globally, with tuberculous pericarditis being most common in developing countries. Other important etiologies include fungal, autoimmune, autoinflammatory, neoplastic (both benign and malignant), immunotherapy-related, radiation therapy-induced, metabolic, postcardiac injury, postoperative, and postprocedural causes. Improved understanding of the immune pathophysiological pathways has led to identification and reclassification of some idiopathic pericarditis cases into autoinflammatory etiologies, including immunoglobulin G (IgG)4-related pericarditis, tumour necrosis factor receptor-associated periodic syndrome (TRAPS), and familial Mediterranean fever in the current era. Contemporary advances in percutaneous cardiac interventions and the recent COVID-19 pandemic have also resulted in changes in the epidemiology of pericardial diseases. Further research is needed to improve our understanding of the etiologies of pericarditis, using the assistance of contemporary advanced imaging techniques and laboratory testing. Careful consideration of the range of potential causes and local epidemiologic patterns of causality are important for the optimization of diagnostic and therapeutic approaches.

Pulmonary artery pulsatility index in patients with tricuspid valve regurgitation: a simple non-invasive tool for risk stratification.

AIMS: Tricuspid valve regurgitation (TR) is a common valvular disease associated with increased mortality. There is a need for tools to assess the interaction between the pulmonary artery (PA) circulation and the right ventricle in patients with TR and to investigate their association with outcomes. The pulmonary artery pulsatility index (PAPi) has emerged as a haemodynamic risk predictor in left heart disease and pulmonary hypertension (PH). Whether PAPi discriminates risk in unselected patients with greater than or equal to moderate TR is unknown. METHODS AND RESULTS: In 5079 patients with greater than or equal to moderate TR (regardless of aetiology) and PA systolic and diastolic pressures measured on their first echocardiogram, we compared all-cause mortality at 5 years based on the presence or absence of PH and PAPi levels. A total of 2741 (54%) patients had PH. The median PAPi was 3.0 (IQR 1.9, 4.4). Both the presence of PH and decreasing levels of PAPi were associated with larger right ventricles, worse right ventricular systolic function, higher NT-pro BNP levels, greater degrees of right heart failure, and worse survival. In a subset of patients who had an echo and right heart catheterization within 24 h, the correlation of non-invasive to invasive PA pressures and PAPi levels was very good (r = 0.76). CONCLUSION: In patients with greater than or equal to moderate TR with and without PH, lower PAPi is associated with right ventricular dysfunction, right heart failure, and worse survival. Incorporating PA pressure and PAPi may help stratify disease severity in patients with greater than or equal to moderate TR regardless of aetiology.

Developing an Echocardiography-Based, Automatic Deep Learning Framework for the Differentiation of Increased Left Ventricular Wall Thickness Etiologies.

AIMS: Increased left ventricular (LV) wall thickness is frequently encountered in transthoracic echocardiography (TTE). While accurate and early diagnosis is clinically important, given the differences in available therapeutic options and prognosis, an extensive workup is often required to establish the diagnosis. We propose the first echo-based, automated deep learning model with a fusion architecture to facilitate the evaluation and diagnosis of increased left ventricular (LV) wall thickness. METHODS AND RESULTS: Patients with an established diagnosis of increased LV wall thickness (hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), and hypertensive heart disease (HTN)/others) between 1/2015 and 11/2019 at Mayo Clinic Arizona were identified. The cohort was divided into 80%/10%/10% for training, validation, and testing sets, respectively. Six baseline TTE views were used to optimize a pre-trained InceptionResnetV2 model. Each model output was used to train a meta-learner under a fusion architecture. Model performance was assessed by multiclass area under the receiver operating characteristic curve (AUROC). A total of 586 patients were used for the final analysis (194 HCM, 201 CA, and 191 HTN/others). The mean age was 55.0 years, and 57.8% were male. Among the individual view-dependent models, the apical 4-chamber model had the best performance (AUROC: HCM: 0.94, CA: 0.73, and HTN/other: 0.87). The final fusion model outperformed all the view-dependent models (AUROC: HCM: 0.93, CA: 0.90, and HTN/other: 0.92). CONCLUSION: The echo-based InceptionResnetV2 fusion model can accurately classify the main etiologies of increased LV wall thickness and can facilitate the process of diagnosis and workup.