Background: Artificial intelligence, particularly deep learning (DL), has immense potential to improve the interpretation of transthoracic echocardiography (TTE). Mitral regurgitation (MR) is the most common valvular heart disease and presents unique challenges for DL, including the integration of multiple video-level assessments into a final study-level classification.
Methods: A novel DL system was developed to intake complete TTEs, identify color MR Doppler videos, and determine MR severity on a 4-step ordinal scale (none/trace, mild, moderate, and severe) using the reading cardiologist as a reference standard. This DL system was tested in internal and external test sets with performance assessed by agreement with the reading cardiologist, weighted κ, and area under the receiver-operating characteristic curve for binary classification of both moderate or greater and severe MR. In addition to the primary 4-step model, a 6-step MR assessment model was studied with the addition of the intermediate MR classes of mild-moderate and moderate-severe with performance assessed by both exact agreement and ±1 step agreement with the clinical MR interpretation.
Results: A total of 61 689 TTEs were split into train (n=43 811), validation (n=8891), and internal test (n=8987) sets with an additional external test set of 8208 TTEs. The model had high performance in MR classification in internal (exact accuracy, 82%; κ=0.84; area under the receiver-operating characteristic curve, 0.98 for moderate or greater MR) and external test sets (exact accuracy, 79%; κ=0.80; area under the receiver-operating characteristic curve, 0.98 for moderate or greater MR). Most (63% internal and 66% external) misclassification disagreements were between none/trace and mild MR. MR classification accuracy was slightly higher using multiple TTE views (accuracy, 82%) than with only apical 4-chamber views (accuracy, 80%). In subset analyses, the model was accurate in the classification of both primary and secondary MR with slightly lower performance in cases of eccentric MR. In the analysis of the 6-step classification system, the exact accuracy was 80% and 76% with a ±1 step agreement of 99% and 98% in the internal and external test set, respectively.
Conclusions: This end-to-end DL system can intake entire echocardiogram studies to accurately classify MR severity and may be useful in helping clinicians refine MR assessments.
To achieve cardiovascular health (CVH) equity in the United States, an understanding of the social and structural factors that contribute to differences and disparities in health is necessary. The Asian American population is the fastest-growing racial group in the United States but remains persistently underrepresented in health research. There is heterogeneity in how individual Asian American ethnic groups experience CVH and cardiovascular disease outcomes, with certain ethnic groups experiencing a higher burden of adverse social conditions, disproportionately high burden of suboptimal CVH, or excess adverse cardiovascular disease outcomes. In this scientific statement, upstream structural and social determinants that influence CVH in the Asian American population are highlighted, with particular emphasis on the role of social determinants of health across disaggregated Asian American ethnic groups. Key social determinants that operate in Asian American communities include socioeconomic position, immigration and nativity, social and physical environments, food and nutrition access, and health system-level factors. The role of underlying structural factors such as health, social, and economic policies and structural racism is also discussed in the context of CVH in Asian Americans. To improve individual-, community-, and population-level CVH and to reduce CVH disparities in Asian American ethnic subgroups, multilevel interventions that address adverse structural and social determinants are critical to achieve CVH equity for the Asian American population. Critical research gaps for the Asian American population are given, along with recommendations for strategic approaches to investigate social determinants of health and intervene to reduce health disparities in these communities.
Recent advances in therapy and the promulgation of multidisciplinary pulmonary embolism teams show great promise to improve management and outcomes of acute pulmonary embolism (PE). However, the absence of randomized evidence and lack of consensus leads to tremendous variations in treatment and compromises the wide implementation of new innovations. Moreover, the changing landscape of health care, where quality, cost, and accountability are increasingly relevant, dictates that a broad spectrum of outcomes of care must be routinely monitored to fully capture the impact of modern PE treatment. We set out to standardize data collection in patients with PE undergoing evaluation and treatment, and thus establish the foundation for an expanding evidence base that will address gaps in evidence and inform future care for acute PE. To do so, >100 international PE thought leaders convened in Washington, DC, in April 2022 to form the Pulmonary Embolism Research Collaborative. Participants included physician experts, key members of the US Food and Drug Administration, patient representatives, and industry leaders. Recognizing the multidisciplinary nature of PE care, the Pulmonary Embolism Research Collaborative was created with representative experts from stakeholder medical subspecialties, including cardiology, pulmonology, vascular medicine, critical care, hematology, cardiac surgery, emergency medicine, hospital medicine, and pharmacology. A list of critical evidence gaps was composed with a matching comprehensive set of standardized data elements; these data points will provide a foundation for productive research, knowledge enhancement, and advancement of clinical care within the field of acute PE, and contribute to answering urgent unmet needs in PE management. Evidence produced through the Pulmonary Embolism Research Collaborative, as it is applied to data collection, promises to provide crucial knowledge that will ultimately produce a robust evidence base that will lead to standardization and harmonization of PE management and improved outcomes.
Background: The effect of myocardial infarction (MI) on life expectancy is difficult to study because the prevalence of MI hinders direct comparison with the life expectancy of the general population. We sought to assess this in relation to age, sex, and left ventricular ejection fraction (LVEF) by comparing individuals with MI with matched comparators without previous MI.
Methods: We included patients with a first MI between 1991 and 2022 from the nationwide SWEDEHEART registry (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies), each matched with up to 5 comparators on age, sex, and region of residence. Flexible parametric survival models were used to estimate excess mortality and mean loss of life expectancy (LOLE) depending on index year, age, sex, and LVEF, and adjusted for differences in characteristics.
Results: A total of 335 748 cases were matched to 1 625 396 comparators. A higher LOLE was observed in younger individuals, women, and those with reduced LVEF (<50%). In 2022, the unadjusted and adjusted mean LOLE spanned from 11.1 and 9.5 years in 50-year-old women with reduced LVEF to 5 and 3.7 months in 80-year-old men with preserved LVEF. Between 1992 and 2022, the adjusted mean LOLE decreased by 36% to 55%: from 4.4 to 2.0 years and from 3.3 to 1.9 years in 50-year-old women and men, respectively, and from 1.7 to 1.0 years and from 1.4 to 0.9 years in 80-year-old women and men, respectively.
Conclusions: LOLE is higher in younger individuals, women, and those with reduced LVEF, but is attenuated when adjusting for comorbidities and risk factors. Advances in MI treatment during the past 30 years have almost halved LOLE, with no clear sign of leveling off to a plateau.