Valvular heart disease (VHD) is a growing public health concern, yet over half of cases remain undiagnosed due to late symptom onset, limited public awareness, and low sensitivity of traditional stethoscope-based screening. Current AI-enabled tools rely on murmur detection as a proxy for VHD but lack sensitivity for common subtypes like mitral regurgitation and are limited by small datasets. This study presents a novel neural network that directly predicts clinically significant VHD from stethoscope recordings, trained using echocardiographic targets rather than heart murmur labels. A diverse dataset of 1767 patients across UK primary care and hospital settings was developed, combining stethoscope recordings with echocardiographic labels. The trained recurrent neural network achieved an AUROC of 0.83, outperforming general practitioners and demonstrating exceptional sensitivity for severe aortic stenosis (98%) and severe mitral regurgitation (94%). This algorithm shows promise as a scalable, low-cost screening tool, enabling earlier diagnosis and timely referral for intervention. This research was registered with ClinicalTrials.gov (CAIS: NCT04445012 registered on 2020-06-21, DUO-EF: NCT04601415 registered on 2020-10-19).

The right ventricle (RV) and tricuspid valve (TV) have historically been neglected in cardiovascular research and clinical practice. However, growing evidence has established RV dysfunction and tricuspid regurgitation (TR) as independent predictors of mortality and morbidity across a range of cardiac pathologies. The emergence of transcatheter tricuspid valve interventions (TTVI) has further emphasized the need for accurate and reproducible assessment of RV function in the presence of TR. To provide a comprehensive review of current echocardiographic approaches of RV assessment in the context of TR. This review highlights their strengths, limitations, and clinical relevance. We focus particularly on advanced imaging modalities and their role in patient selection and outcome prediction following TTVI. We review contemporary literature on RV anatomy and pathophysiology and critically evaluate echocardiographic modalities, including two-dimensional (2D) indices, three-dimensional (3D) volumetry, strain imaging, and noninvasive RV-pulmonary artery (PA) coupling surrogates, for their role in diagnosing RV dysfunction, guiding risk stratification, and predicting clinical outcomes in patients considered for TTVI. Conventional echocardiographic measures of RV function such as tricuspid annular plane systolic excursion (TAPSE) and fractional area change (FAC) are limited by geometric assumptions, regional motion bias, and load dependence. Emerging echocardiographic techniques, including 3D imaging, free wall longitudinal strain (FWLS), effective RV ejection fraction (eRVEF), and RV-PA coupling indices, demonstrate improved accuracy and prognostic utility. In patients with TR, progressive RV remodeling, altered contraction patterns, and altered ventricular interdependence contribute to clinical decompensation. While TTVI provides hemodynamic and symptomatic benefit, it can also reveal latent RV dysfunction, underscoring the importance of thorough pre- and postprocedural evaluation. Accurate assessment of RV structure and function is critical for optimal management of TR, particularly in candidates for TTVI. A multiparametric echocardiographic strategy that integrates advanced imaging techniques with functional indices provides a more complete characterization of RV performance, supports procedural planning, and improves risk stratification. Future research should aim to establish TR-specific thresholds, validate emerging functional markers, and develop standardized, evidence-based algorithms to guide clinical decision-making.

Abstract Background Artificial intelligence (AI) measurement has the potential to transform cardiac imaging, but what if it makes a mistake? Can AI also highlight when it is most likely to have mismeasured and can experts improve these measurements? As readers become increasingly reliant on automated measures for echocardiographic analysis, we need confidence that the software will measure accurately, and also flag where human oversight is needed. Purpose To develop and test an open, scientific, machine-learning method of validly judging the level of certainty in an AI measurement, to enable human oversight to focus on the measurements most likely to be incorrect. Methods The heatmap altitude (MHA) at the measurement point is widely used as an automatic index of predicted reliability of an AI measurement. This is derived from the raw neural network output to give a level of confidence between 0 and 1 in each measurement that the AI makes. We test this not only against the AI’s difference from the expert consensus, but also, through the use of multiple experts per case, against individual experts difference from the consensus using the Unity UK Echocardiography AI collaboratives dataset of 200 parasternal long-axis images. Each image was labelled with key points for the aortic annulus, sinus and sinotubular junction, and proximal ascending aorta dimensions by 10 experts. The mean expert consensus measure was obtained, and then the median deviation of the AI and other experts calculated. Results The heatmap altitude was skewed with median 0.822, IQR 0.793 to 0.841, 10th to 90th percentile 0.734 to 0.852. Images were grouped by decile, from the 20 images with lowest confidence to the 20 with the highest (Figure 1). Both the AI and expert error (deviation from expert consensus) was greatest in the lowest confidence decile and progressively lower in each decile of greater confidence (p<0.001 for trend). In the better 9 deciles, the AI error was smaller than the human error (p<0.001); in the worst decile it was equivalent (p =NS). Conclusion The fully automated and open-source MHA value usefully quantifies the reliability of an AI measurement. This could be used to target human oversight where it is most needed, but care must be taken as experts may find the images similarly difficult.

Impairment of left ventricular (LV) diastolic function is common amongst those with left heart disease and is associated with significant morbidity. Given that, in simple terms, the ventricle can only eject the volume with which it fills and that approximately one half of hospitalisations for heart failure (HF) are in those with normal/’preserved’ left ventricular ejection fraction (HFpEF) (Bianco et al. in JACC Cardiovasc Imaging. 13:258–271, 2020. 10.1016/j.jcmg.2018.12.035), where abnormalities of ventricular filling are the cause of symptoms, it is clear that the assessment of left ventricular diastolic function (LVDF) is crucial for understanding global cardiac function and for identifying the wider effects of disease processes. Invasive methods of measuring LV relaxation and filling pressures are considered the gold-standard for investigating diastolic function. However, the high temporal resolution of trans-thoracic echocardiography (TTE) with widely validated and reproducible measures available at the patient’s bedside and without the need for invasive procedures involving ionising radiation have established echocardiography as the primary imaging modality. The comprehensive assessment of LVDF is therefore a fundamental element of the standard TTE (Robinson et al. in Echo Res Pract7:G59–G93, 2020. 10.1530/ERP-20-0026). However, the echocardiographic assessment of diastolic function is complex. In the broadest and most basic terms, ventricular diastole comprises an early filling phase when blood is drawn, by suction, into the ventricle as it rapidly recoils and lengthens following the preceding systolic contraction and shortening. This is followed in late diastole by distension of the compliant LV when atrial contraction actively contributes to ventricular filling. When LVDF is normal, ventricular filling is achieved at low pressure both at rest and during exertion. However, this basic description merely summarises the complex physiology that enables the diastolic process and defines it according to the mechanical method by which the ventricles fill, overlooking the myocardial function, properties of chamber compliance and pressure differentials that determine the capacity for LV filling. Unlike ventricular systolic function where single parameters are utilised to define myocardial performance (LV ejection fraction (LVEF) and Global Longitudinal Strain (GLS)), the assessment of diastolic function relies on the interpretation of multiple myocardial and blood-flow velocity parameters, along with left atrial (LA) size and function, in order to diagnose the presence and degree of impairment. The echocardiographic assessment of diastolic function is therefore multifaceted and complex, requiring an algorithmic approach that incorporates parameters of myocardial relaxation/recoil, chamber compliance and function under variable loading conditions and the intra-cavity pressures under which these processes occur. This guideline outlines a structured approach to the assessment of diastolic function and includes recommendations for the assessment of LV relaxation and filling pressures. Non-routine echocardiographic measures are described alongside guidance for application in specific circumstances. Provocative methods for revealing increased filling pressure on exertion are described and novel and emerging modalities considered. For rapid access to the core recommendations of the diastolic guideline, a quick-reference guide (additional file 1) accompanies the main guideline document. This describes in very brief detail the diastolic investigation in each patient group and includes all algorithms and core reference tables.

Abstract Background Surveillance of aortic dimensions requires reproducible measurements, and knowledge of the rate of progression in those with dilatation. Purpose To develop an open machine-learning method for measuring the aortic root and proximal ascending aorta on echocardiograms, validate it through a multi-expert panel from the Unity UK Echocardiography AI Collaborative, and applying the AI to derive the rate of progression in historical databases. Methods The neural network was trained on 1478 parasternal long-axis images. Each image was labelled with key points for the aortic annulus, sinus and sinotubular junction, and proximal ascending aorta dimensions. Labels accommodated inner-to-inner and leading-to-leading conventions. For validation, the end-diastolic and mid-systolic images were identified (accommodating different guidelines recommendations) from 100 studies, and 10 expert echocardiographers made the aortic measurements. The consensus of experts defined the reference standard, and the variation between individual experts defined the acceptable variation for the AI. The AI was then applied to 724 echocardiograms of 102 patients under surveillance spanning 12 years (ranging from 4 to 17 scans per patient) without intervention. Training data labels and networks are made freely available on our project website. Results The median absolute deviations between the AI and the expert consensus ranged from 0.06cm to 0.17cm across the 16 measurements. For the individual expert opinions this was from 0.08cm to 0.19cm. For 7/16 measurements, there was no statistically significant difference between the AI deviation and the individual-expert deviation. The AI’s confidence level was a useful indicator of the reliability of the AI measure. For the top 9 deciles of AI confidence level, the AI measurements had significantly smaller deviations than the individual experts. The rates of progression averaged across the three aortic measurements was 2.0 mm/decade (95% CI 1.0 to 3.0; p =0.007). Conclusion Expert consensus can be used to both define the reference standard and also the range of acceptable deviation from consensus. Overall, the AI performs similarly to experts and more importantly, provides an automated estimate of confidence. The validated AI is consistent over time, and when applied retrospectively to patients with dilated aortic roots, found the rate of progression was more rapid than in previously reported healthy patients. This approach could be applied generally in developing medical AI to not only make reproducible measurements but to automate large scale longitudinal studies with little input, and ultimately serves as a useful research and clinical tool.Figure: AI(red) vs Expert (blue)Table: AIDeviations from Consensus

BACKGROUND Global longitudinal strain (GLS) is reported to be more reproducible and prognostic than ejection fraction. Automated, transparent methods may increase trust and uptake. OBJECTIVES The authors developed open machine-learning-based GLS methodology and validate it using multiexpert consensus from the Unity UK Echocardiography AI Collaborative. METHODS We trained a multi-image neural network (Unity-GLS) to identify annulus, apex, and endocardial curve on 6,819 apical 4-, 2-, and 3-chamber images. The external validation dataset comprised those 3 views from 100 echocardiograms. End-systolic and-diastolic frames were each labelled by 11 experts to form consensus tracings and points. They also ordered the echocardiograms by visual grading of longitudinal function. One expert calculated global strain using 2 proprietary packages. RESULTS The median GLS, averaged across the 11 individual experts, was-16.1 (IQR:-19.3 to-12.5). Using each case's expert consensus measurement as the reference standard, individual expert measurements had a median absolute error of 2.00 GLS units. In comparison, the errors of the machine methods were: Unity-GLS 1.3, proprietary A 2.5, proprietary B 2.2. The correlations with the expert consensus values were for individual experts 0.85, Unity-GLS 0.91, proprietary A 0.73, proprietary B 0.79. Using the multiexpert visual ranking as the reference, individual expert strain measurements found a median rank correlation of 0.72, Unity-GLS 0.77, proprietary A 0.70, and proprietary B 0.74. CONCLUSIONS Our open-source approach to calculating GLS agrees with experts' consensus as strongly as the individual expert measurements and proprietary machine solutions. The training data, code, and trained networks are freely available online. (JACC Cardiovasc Imaging 2024;17:865-876) (c) 2024 The Authors. Published by Elsevier on behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Left atrial appendage occlusion (LAAO) has emerged as a highly effective alternative to oral anticoagulation for stroke prevention in patients with non-valvular atrial fibrillation. Precise pre-procedural planning and meticulous post-procedural follow-up are essential for achieving successful LAAO outcomes. This review explores the latest advancements in three-dimensional (3D) transoesophageal echocardiography (TOE) and computed tomography (CT) imaging modalities, which have considerably improved the planning, intra-procedural guidance, and follow-up processes for LAAO interventions. Innovations in 3D TOE and CT imaging have transformed the approach to LAAO by providing a more detailed and accurate assessment of the left atrial appendage, enabling clinicians to acquire comprehensive anatomical and morphological information, crucial for optimising device selection and positioning, thus reducing the risk of complications and enhancing the overall safety and efficacy of the procedure. Post-procedurally, CT and TOE imaging are invaluable in the monitoring of patients, ensuring that the device is correctly positioned and functioning as intended. Early detection of any complications (e.g., device-related thrombus and peri-device leaks) can help to risk-stratify patient at increased risk of stroke and initiate timely interventions, thereby improving long-term outcomes for patients.

Ultrasound contrast agents (UCAs) have a well-established role in clinical cardiology. Contrast echocardiography has evolved into a routine technique through the establishment of contrast protocols, an excellent safety profile, and clinical guidelines which highlight the incremental prognostic utility of contrast enhanced echocardiography. This document aims to provide practical guidance on the safe and effective use of contrast; reviews the role of individual staff groups; and training requirements to facilitate its routine use in the echocardiography laboratory.

Infective endocarditis (IE) carries a high risk of morbidity and mortality. Timely diagnosis, effective treatment and prompt recognition of complications are essential to favourable patient outcomes. A collaborative, multidisciplinary team approach to the management of IE has been shown to improve prognosis. However, the clinical heterogeneity of IE and atypical presentations pose challenges to the endocarditis team. We present a case highlighting the role of valve histopathology in suspected IE, where there may be diagnostic uncertainty.

•Stroke is a significant complication of transcatheter aortic valve replacement.•Computed tomography can quantify aortic valve calcification prior to intervention.•Aortic valve calcium on CT was associated with post-TAVR stroke.•This could inform discussions of individual patient stroke risk prior to TAVR.

Summary: Background: Most patients who have heart failure with a reduced ejection fraction, when left ventricular ejection fraction (LVEF) is 40% or lower, are diagnosed in hospital. This is despite previous presentations to primary care with symptoms. We aimed to test an artificial intelligence (AI) algorithm applied to a single-lead ECG, recorded during ECG-enabled stethoscope examination, to validate a potential point-of-care screening tool for LVEF of 40% or lower. Methods: We conducted an observational, prospective, multicentre study of a convolutional neural network (known as AI-ECG) that was previously validated for the detection of reduced LVEF using 12-lead ECG as input. We used AI-ECG retrained to interpret single-lead ECG input alone. Patients (aged ≥18 years) attending for transthoracic echocardiogram in London (UK) were recruited. All participants had 15 s of supine, single-lead ECG recorded at the four standard anatomical positions for cardiac auscultation, plus one handheld position, using an ECG-enabled stethoscope. Transthoracic echocardiogram-derived percentage LVEF was used as ground truth. The primary outcome was performance of AI-ECG at classifying reduced LVEF (LVEF ≤40%), measured using metrics including the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity, with two-sided 95% CIs. The primary outcome was reported for each position individually and with an optimal combination of AI-ECG outputs (interval range 0–1) from two positions using a rule-based approach and several classification models. This study is registered with ClinicalTrials.gov, NCT04601415. Findings: Between Feb 6 and May 27, 2021, we recruited 1050 patients (mean age 62 years [SD 17·4], 535 [51%] male, 432 [41%] non-White). 945 (90%) had an ejection fraction of at least 40%, and 105 (10%) had an ejection fraction of 40% or lower. Across all positions, ECGs were most frequently of adequate quality for AI-ECG interpretation at the pulmonary position (979 [93·3%] of 1050). Quality was lowest for the aortic position (846 [80·6%]). AI-ECG performed best at the pulmonary valve position (p=0·02), with an AUROC of 0·85 (95% CI 0·81–0·89), sensitivity of 84·8% (76·2–91·3), and specificity of 69·5% (66·4–72·6). Diagnostic odds ratios did not differ by age, sex, or non-White ethnicity. Taking the optimal combination of two positions (pulmonary and handheld positions), the rule-based approach resulted in an AUROC of 0·85 (0·81–0·89), sensitivity of 82·7% (72·7–90·2), and specificity of 79·9% (77·0–82·6). Using AI-ECG outputs from these two positions, a weighted logistic regression with l2 regularisation resulted in an AUROC of 0·91 (0·88–0·95), sensitivity of 91·9% (78·1–98·3), and specificity of 80·2% (75·5–84·3). Interpretation: A deep learning system applied to single-lead ECGs acquired during a routine examination with an ECG-enabled stethoscope can detect LVEF of 40% or lower. These findings highlight the potential for inexpensive, non-invasive, workflow-adapted, point-of-care screening, for earlier diagnosis and prognostically beneficial treatment. Funding: NHS Accelerated Access Collaborative, NHSX, and the National Institute for Health Research.

Tricuspid valve pathophysiology is not well-understood. Emergence of novel transcatheter tricuspid therapies has fueled the requirements for improved imaging visualization techniques and interventional imaging physician skillsets in guiding these complex transcatheter procedures. There is growing understanding on the clinical significance of tricuspid regurgitation which expanded the interest for percutaneous tricuspid valve interventions. The present review concentrates on three essential aspects of tricuspid valve pathophysiology: anatomical considerations for tricuspid interventions, optimal timing of tricuspid interventions by imaging guidance, and the role of interventional imaging physicians’ skillset and knowledge in this field.

The subspecialty of cardio-oncology aims to reduce cardiovascular morbidity and mortality in patients with cancer or following cancer treatment. Cancer therapy can lead to a variety of cardiovascular complications, including left ventricular systolic dysfunction, pericardial disease, and valvular heart disease. Echocardiography is a key diagnostic imaging tool in the diagnosis and surveillance for many of these complications. The baseline assessment and subsequent surveillance of patients undergoing treatment with anthracyclines and/or human epidermal growth factor (EGF) receptor (HER) 2-positive targeted treatment (e.g. trastuzumab and pertuzumab) form a significant proportion of cardio-oncology patients undergoing echocardiography. This guideline from the British Society of Echocardiography and British Cardio-Oncology Society outlines a protocol for baseline and surveillance echocardiography of patients undergoing treatment with anthracyclines and/or trastuzumab. The methodology for acquisition of images and the advantages and disadvantages of techniques are discussed. Echocardiographic definitions for considering cancer therapeutics-related cardiac dysfunction are also presented.

The COVID-19 pandemic has altered our approach to inpatient echocardiography delivery. There is now a greater focus to address key clinical questions likely to make an immediate impact in management, particularly during the period of widespread infection. Handheld echocardiography (HHE) can be used as a first-line assessment tool, limiting scanning time and exposure to high viral load. This article describes a potential role for HHE during a pandemic. We propose a protocol with a reporting template for a focused core dataset necessary in delivering an acute echocardiography service in the setting of a highly contagious disease, minimising risk to the operator. We cover the scenarios typically encountered in the acute cardiology setting and how an expert trained echocardiography team can identify such pathologies using a limited imaging format and include cardiac presentations encountered in those patients acutely unwell with COVID-19.

BACKGROUND:Infective endocarditis (IE) is challenging to manage in the COVID-19 lockdown period, in part given its reliance on echocardiography for diagnosis and management and the associated virus transmission risks to patients and healthcare workers. This study assesses utilisation of the endocarditis team (ET) in limiting routine echocardiography, especially transoesophageal echocardiography (TOE), in patients with suspected IE, and explores the effect on clinical outcomes.METHODS:All patients discussed at the ET meeting at Imperial College Healthcare NHS Trust during the first lockdown in the UK (23 March to 8 July 2020) were prospectively included and analysed in this observational study.RESULTS:In total, 38 patients were referred for ET review (71% male, median age 54 [interquartile range 48, 65.5] years). At the time of ET discussion, 21% had no echo imaging, 16% had point-of-care ultrasound only, and 63% had formal TTE. In total, only 16% underwent TOE. The ability of echocardiography, in those where it was performed, to affect IE diagnosis according to the Modified Duke Criteria was significant (p=0.0099); however, sensitivity was not affected. All-cause mortality was 17% at 30 days and 25% at 12 months from ET discussion in those with confirmed IE.CONCLUSION:Limiting echocardiography in patients with a low pretest probability (not probable or definite IE according to the Modified Duke Criteria) did not affect the diagnostic ability of the Modified Duke Criteria to rule out IE in this small study. Moreover, restricting nonessential echocardiography, and importantly TOE, in patients with suspected IE through use of the ET did not impact all-cause mortality.

This editorial refers to 'Prevalence and extent of mitral annular disjunction in structurally normal hearts: comprehensive three-dimensional analysis using cardiac computed tomography' H. Toh et al. pp. 614--22. In 1986, Hutchins et al.1 examined 900 post-mortem hearts, where 25 had typical features of mitral valve prolapse (MVP). The mitral annulus was described. Twenty-three of these MVP hearts showed evidence of separation of atrial wall-mitral valve leaflet junction with the ventricular myocardium. Only 42 of 875 non-MVP hearts showed a similar finding. This apparent anomaly was coined mitral annulus disjunction (MAD). Since the latter group were considerably younger, the authors postulated that over time the hypermobility of the mitral annulus may be linked with the progression of myxomatous valve disease. It is worth noting only one block was taken for histological examination from the obtuse margin of the annulus from each normal heart. A number of studies, both post-mortem...

Abstract Background/Introduction Artificial intelligence (AI) applied to 12-lead ECG can identify left ventricular ejection fraction (EF) ≤35% with a sensitivity and specificity of 86.3% and 85.7%, respectively. Whether AI algorithms trained on 12-lead can accurately predict EF from single-lead ECGs (recorded by a smart stethoscope) remains unknown. This could facilitate point-of-care screening for low EF during routine clinical examination. Purpose First independent multicentre real-world UK National Health Service (NHS) prospective validation of 12-lead-ECG-trained AI algorithm applied to single-lead ECG recorded by a smart stethoscope, with AI algorithm tuned to detect EF ≤40%. Methods Prospective recruitment of unselected patients attending for echocardiography across six urban NHS hospital sites (UK). In addition to transthoracic echocardiogram (routine care), all participants had 15 seconds of supine, single-lead ECG recorded at six different positions (figure), encompassing standard anatomical positions for cardiac auscultation. A convolutional neural network (CNN) previously trained on 35,970 independent pairings of 12-lead-ECG and echocardiograms was retrained to use the single-lead ECG as input. Accuracy of CNN detection of low EF (binary ≤40%) is reported at a threshold of 0.5 against gold-standard; echo-determined percentage EF. Results Among 353 patients recruited (mean age 63±17; 58% male, 43.1% non-white), 309 (87.5%) had an EF >40%, and 44 (12.5%) had EF ≤40%. The best single recording position in isolation was position 3 (sensitivity 57.9% [42.2–73.6], specificity 86.3% [82.2–90.3]). Taking any of the six positions performed during the examination as predicting EF ≤40%, this achieved a sensitivity of 81.2% and specificity of 61.5%. Conclusion(s) In this first prospective multicentre validation study the retrained AI algorithm reliably detected low EF from single-lead ECGs acquired using a novel ECG-enabled stethoscope in standard auscultation positions. The ability to identify patients with possible low EF during routine physical examination addresses a significant unmet clinical need in point-of-care ruling in/out of heart failure, and has potential to provide broader population-level screening for asymptomatic cardiovascular disease. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Institute of Health Research, Accelerated Access Collaborative & NHSX: Artificial Intelligence in Health & Social Care Award

Background: requires training and validation to standards expected of humans. We developed an online platform and established the Unity Collaborative to build a dataset of expertise from 17 hospitals for training, validation, and standardization of such techniques. Methods: The training dataset consisted of 2056 individual frames drawn at random from 1265 parasternal long-axis video-loops of patients undergoing clinical echocardiography in 2015 to 2016. Nine experts labeled these images using our online platform. From this, we trained a convolutional neural network to identify keypoints. Subsequently, 13 experts labeled a validation dataset of the end-systolic and end-diastolic frame from 100 new video-loops, twice each. The 26-opinion consensus was used as the reference standard. The primary outcome was precision SD, the SD of the differences between AI measurement and expert consensus. Results: In the validation dataset, the AI's precision SD for left ventricular internal dimension was 3.5 mm. For context, precision SD of individual expert measurements against the expert consensus was 4.4 mm. Intraclass correlation coefficient between AI and expert consensus was 0.926 (95% CI, 0.904-0.944), compared with 0.817 (0.778-0.954) between individual experts and expert consensus. For interventricular septum thickness, precision SD was 1.8 mm for AI (intraclass correlation coefficient, 0.809; 0.729-0.967), versus 2.0 mm for individuals (intraclass correlation coefficient, 0.641; 0.568-0.716). For posterior wall thickness, precision SD was 1.4 mm for AI (intraclass correlation coefficient, 0.535 [95% CI, 0.379-0.661]), versus 2.2 mm for individuals (0.366 [0.288-0.462]). We present all images and annotations. This highlights challenging cases, including poor image quality and tapered ventricles. Conclusions: Experts at multiple institutions successfully cooperated to build a collaborative AI. This performed as well as individual experts. Future echocardiographic AI research should use a consensus of experts as a reference. Our collaborative welcomes new partners who share our commitment to publish all methods, code, annotations, and results openly.

Percutaneous closure of paravalvar leaks (PVLs) was once only performed in extreme or non-surgical risk cases not suitable for redo-surgery with tissue or mechanical valves. This technique is now the treatment of choice with long term outcomes that are better than redo operations.123 As interventionalists become more familiar with using PVL devices, more off label device use has been reported in non-surgical cases involving complex native mitral valve regurgitation (NVMR). In this review, we appraise the literature regarding percutaneous treatment of paravalvar leaks and more recently esoteric off label use. We also describe two recent challenging cases where percutaneous devices were used to treat severe leaks that developed as a sequela of previous infective endocarditis. We also provide specific recommendations regarding periprocedural strategy and focus on the importance of device choice to provide an optimal outcome.