Echocardiography

BackgroundDiagnosing cardiac amyloidosis (CA) on echocardiography can be challenging due to the imaging overlap between CA and more prevalent causes of a hypertrophic phenotype. This study sought to (1) evaluate the performance of artificial-intelligence (AI) derived measurements incorporated into the established multiparametric echocardiographic scoring system to detect CA; (2) develop and validate an AI-based deep-learning model for video-based detection of CA on echocardiography.MethodsThe study population comprised 5776 patients (CA, 2756; controls, 3020). The training data set included patients from the UK National Amyloidosis Center and Taiwan MacKay Memorial Hospital (CA, 2241; controls, 2130). External test data sets were obtained from the US Duke University Health System (CA, 334; left ventricular hypertrophy controls, 668) and Japan National Cerebral and Cardiovascular Center (CA, 181; left ventricular hypertrophy controls, 222).ResultsThe multiparametric echocardiographic score computed using AI-derived measurements achieved an accuracy of 79.5% (sensitivity, 75.4%; specificity, 81.5%) in the United States cohort and 79.7% (sensitivity, 81.6%; specificity, 78.1%) in the Japan cohort. The deep-learning model demonstrated accuracies of 96.2% (sensitivity, 96.8%; specificity, 95.7%) and 95.8% (sensitivity, 97.3%; specificity, 94.3%) in the internal validation and internal test sets, respectively. External validation of the deep-learning model showed accuracies of 87.5% (sensitivity, 86.6%; specificity, 87.9%) in the United States and 88.4% (sensitivity, 92.3%; specificity, 85.3%) in the Japanese cohort. Subgroup analysis demonstrated that the deep-learning model showed robust discrimination of CA from other hypertrophic phenocopies: CA versus hypertension (area under the curve [AUC], 0.92 [95% CI, 0.91-0.94]), CA versus hypertrophic cardiomyopathy (AUC, 0.91 [95% CI, 0.87-0.94]), CA versus aortic stenosis (AUC, 0.93 [95% CI, 0.90-0.95]), CA versus chronic kidney disease (AUC, 0.93 [95% CI, 0.91-0.95]). The deep-learning model was able to classify a greater proportion of patients compared with the AI-derived multiparametric echocardiographic score and achieved superior diagnostic accuracy (AUC, 0.93 [95% CI, 0.91-0.95] versus AUC, 0.88 [95% CI, 0.85-0.90]; PConclusionsBoth the multiparametric echocardiographic score computed from AI-derived measurements and the fully automated deep-learning model can accurately identify patients with CA in globally diverse cohorts, with the deep-learning model providing superior performance.

Paediatric athletes are not simply 'mini adults'. Most existing recommendations for cardiac screening in paediatric athletes are primarily based on evidence in adults and are designed for adult athletes. Paediatric-specific recommendations are needed due to the specifics of cardiac physiology, maturation and growth, age-related disease expression, modified diagnostic pathways, training adaptations, and to address relevant ethical considerations. This clinical consensus document from the European Association of Preventive Cardiology (EAPC) of the ESC and the Association for European Paediatric and Congenital Cardiology (AEPC) introduces specific advice for paediatric athletes for the first time, based on expert consensus, and where available, data from paediatric athlete populations. Members of the writing group voted anonymously on key advice statements, with ≥80% agreement required for consensus. All advice in this document applies to paediatric athletes aged <16 years, including those under 12 years of age. This document advises that cardiac screening of paediatric athletes with personal and family medical history, physical examination and 12-lead resting electrocardiogram (ECG) should be performed and should start no later than the age of 12 years. Implementing a screening programme requires ensuring the availability of necessary healthcare resources. One transthoracic echocardiogram may be appropriate to identify high-risk structural cardiac diseases not identifiable on ECG, provided appropriate infrastructure for baseline diagnostic assessments is in place. This document also includes suggested definitions of normal, borderline and abnormal ECG findings in paediatric athletes. Detailed advice is provided for further evaluation if suspicious findings are identified on initial tests. This document highlights that further research is required to optimise screening strategies, accurately assess and quantify the risk of sudden cardiac death and provide evidence-based eligibility recommendations for paediatric athletes with cardiac disease. It is also noted that increased opportunities for paediatric sports cardiology training are required to provide adequate medical care for the paediatric athlete population.

AimsIntracardiac echocardiography-based electroanatomical mapping (EAM) improves procedural efficiency and safety in atrial fibrillation (AF) ablation and remains the standard of care. The CARTOSOUND FAM (AI FAM) module uses a deep-learning algorithm that automates left atrial reconstruction without manual contouring. This study aims to evaluate the 1-year outcomes of AI FAM compared to standard-of-care EAM in AF ablation.Methods and resultsThis study included 298 patients undergoing radiofrequency AF ablation between January 2021 and December 2023. Patients treated before January 2023 underwent standard-of-care EAM, while those in 2023 utilized AI FAM-based reconstruction. Baseline demographics, comorbidities, AF type, and medication use were recorded. Procedural characteristics, acute success, complications, and AF recurrence at 1-year follow-up were analysed. Of the 298 patients, 115 underwent mapping with AI FAM and 183 with EAM. Baseline characteristics were comparable. AI FAM reduced mean total procedure time (122.5 ± 23.5 vs. 129.0 ± 30.4 min, P = 0.040) and left atrial (LA) dwell time (78.3 ± 21.45 vs. 87.5 ± 28.2 min, P = 0.001). Acute procedural success was 98.3% in AI FAM vs. 98.9% in EAM with fewer complications observed in the AI FAM group (1 vs. 4). At 1 year, freedom from AF recurrence was comparable (80.0% AI FAM vs. 81.4% EAM at 1 year, LogRank P = 0.610).ConclusionAI FAM was associated with incremental but significant procedural advantages over conventional contouring via reduced total procedure time and LA dwell time, without compromising acute and long-term safety and rhythm control efficacy. AI FAM integration with pulsed field ablation will mark another step towards making AF ablation more streamlined and accessible.