Artificial Intelligence and Echocardiography: A Primer for Cardiac Sonographers.

Artificial intelligence (AI) is emerging as a key component in diagnostic medical imaging, including echocardiography. AI with deep learning has already been used with automated view labeling, measurements, and interpretation. As the development and use of AI in echocardiography increase, potential concerns may be raised by cardiac sonographers and the profession. This report, from a sonographer's perspective, focuses on defining AI, the basics of the technology, identifying some current applications of AI, and how the use of AI may improve patient care in the future.

Myocardial injury and ventricular dysfunction related to training levels among nonelite participants in the Boston marathon.

BACKGROUND: Multiple studies have individually documented cardiac dysfunction and biochemical evidence of cardiac injury after endurance sports; however, convincing associations between the two are lacking. We aimed to determine the associations between the observed transient cardiac dysfunction and biochemical evidence of cardiac injury in amateur participants in endurance sports and to elicit the risk factors for the observed injury and dysfunction. METHODS AND RESULTS: We screened 60 nonelite participants, before and after the 2004 and 2005 Boston Marathons, with echocardiography and serum biomarkers. Echocardiography included conventional measures as well as tissue Doppler-derived strain and strain rate imaging. Biomarkers included cardiac troponin T (cTnT) and N-terminal pro-brain natriuretic peptide (NT-proBNP). All subjects completed the race. Echocardiographic abnormalities after the race included altered diastolic filling, increased pulmonary pressures and right ventricular dimensions, and decreased right ventricular systolic function. At baseline, all had unmeasurable troponin. After the race, > 60% of participants had increased cTnT > 99th percentile of normal (> 0.01 ng/mL), whereas 40% had a cTnT level at or above the decision limit for acute myocardial necrosis (> or = 0.03 ng/mL). After the race, NT-proBNP concentrations increased from 63 (interquartile range [IQR] 21 to 81) pg/mL to 131 (IQR 82 to 193) pg/mL (P<0.001). The increase in biomarkers correlated with post-race diastolic dysfunction, increased pulmonary pressures, and right ventricular dysfunction (right ventricular mid strain, r=-0.70, P<0.001) and inversely with training mileage (r=-0.71, P<0.001). Compared with athletes training > 45 miles/wk, athletes who trained < or = 35 miles/wk demonstrated increased pulmonary pressures, right ventricular dysfunction (mid strain 16+/-5% versus 25+/-4%, P<0.001), myocyte injury (cTnT 0.09 versus < 0.01 ng/mL, P<0.001), and stress (NT-proBNP 182 versus 106 pg/mL, P<0.001). CONCLUSIONS: Completion of a marathon is associated with correlative biochemical and echocardiographic evidence of cardiac dysfunction and injury, and this risk is increased in those participants with less training.

New locus for autosomal dominant mitral valve prolapse on chromosome 13: clinical insights from genetic studies.

BACKGROUND: Mitral valve prolapse (MVP) is a common disorder associated with mitral regurgitation, endocarditis, heart failure, and sudden death. To date, 2 MVP loci have been described, but the defective genes have yet to be discovered. In the present study, we analyzed a large family segregating MVP, and identified a new locus, MMVP3. This study and others have enabled us to explore mitral valve morphological variations of currently uncertain clinical significance. METHODS AND RESULTS: Echocardiograms and blood samples were obtained from 43 individuals who were classified by the extent and pattern of displacement. Genotypic analyses were performed with polymorphic microsatellite markers. Evidence of linkage was obtained on chromosome 13q31.3-q32.1, with a peak nonparametric linkage score of 18.41 (P<0.0007). Multipoint parametric analysis gave a logarithm of odds score of 3.17 at marker D13S132. Of the 6 related individuals with mitral valve morphologies not meeting diagnostic criteria but resembling fully developed forms, 5 carried all or part of the haplotype linked to MVP. CONCLUSIONS: The mapping of a new MVP locus to chromosome 13 confirms the observed genetic heterogeneity and represents an important step toward gene identification. Furthermore, the genetic analysis provides clinical lessons with regard to previously nondiagnostic morphologies. In the familial context, these may represent early expression in gene carriers. Early recognition of gene carriers could potentially enhance the clinical evaluation of patients at risk of full expression, with the ultimate aim of developing interventions to reduce progression.

Usefulness of three-dimensionally guided assessment of mitral stenosis using matrix-array ultrasound.

Two-dimensional (2-D) planimetry is limited by the technical demands, time, and observer variability required to locate the minimal orifice area, limiting the confident clinical reporting of mitral valve area (MVA). In 27 consecutive patients, MVA was determined independently by 2 observers using the conventional 2-D method and a new 3-D-guided method. Using a matrix-array probe, the valve was visualized in a long-axis view and a cursor steered to intersect the leaflet tips and provide a perpendicular short-axis plane viewed side-by-side. Two-dimensional and 3-D-guided methods allowed planimetry in 24 patients. Consistent with better orifice localization, 3-D guidance eliminated the overestimation of internal orifice diameters in the planimetered short-axis view relative to the limiting diameter defined by the long-axis view (for 3-D guidance, 0.73 +/- 0.20 vs 0.73 +/- 0.21 cm, p = 0.98, vs 0.90 +/- 0.27 cm in the 2-D short-axis view, p <0.01). Accordingly, mean values for the smallest orifice area by 3-D guidance were less than by 2-D imaging (1.4 +/- 0.5 vs 1.5 +/- 0.5 cm(2), p <0.01), changing the clinical severity classification in 11 of 24 patients (46%). The 2-D method also overestimated MVA relative to 3-D guidance compared with Doppler pressure halftime and (n = 6) Gorlin areas. Phantom studies verified no differences in resolution for the 2 acquisition modes. Three-dimensional guidance reduced intraobserver variability from 9.8% to 3.8% (SEE 0.14 to 0.06 cm(2), p <0.01) and interobserver variability from 10.6% to 6.1% (SEE 0.15 to 0.09 cm(2), p <0.02). In conclusion, matrix-array technology provides a feasible and highly reproducible direct 3-D-guided method for measuring the limiting mitral orifice area.

Improvement in structural and functional echocardiographic parameters during chronic heart failure therapy guided by natriuretic peptides: mechanistic insights from the ProBNP Outpatient Tailored Chronic Heart Failure (PROTECT) study.

AIMS: We sought to determine if heart failure (HF) care with a goal to lower N-terminal pro B-type natriuretic peptide (NT-proBNP) concentrations, compared with standard of care (SOC) management, is associated with improvement in echocardiographic parameters of cardiac structure and function. METHODS AND RESULTS: Of 151 subjects with HF due to left ventricular systolic dysfunction (LVSD) prospectively randomized to NT-proBNP-guided vs. SOC HF care, 116 had serial echocardiographic data. Endpoints in this echocardiographic study included the relationship between change in NT-proBNP and LV reverse remodelling, as well as associations between biomarker-guided therapy and measures of diastolic function, right ventricular (RV) size and function, estimates of LV filling pressure and RV systolic pressure (RVSP), and the degree of mitral regurgitation (MR). After a mean of 10 months of study procedures, in adjusted analyses, final NT-proBNP concentrations predicted risk of remodelling [hazard ratio (HR) ↑LV end-diastolic volume index = 1.43, 95% confidence interval (CI) 1.10-1.86, P = 0.007; HR ↑LV end-systolic volume index = 1.54, 95% CI 1.10-1.91, P = 0.01; HR ↓LV ejection fraction (LVEF) = 1.53, 905% CI 1.12-1.89, P = 0.02]. In addition to greater improvement in LVEF and reductions in LV volume, compared with SOC, NT-proBNP-guided patients showed significant decreases in the ratio of early transmitral peak velocity to early diastolic peak annular velocity (E/E'), pulmonary vein peak S velocity, RV fractional area change, RVSP, and MR severity. CONCLUSION: NT-proBNP concentrations may serve as a non-invasive indicator of the state of cardiac structure and function in HF due to LVSD. Multiple, prognostically meaningful echocardiographic variables improved more significantly in patients treated with NT-proBNP-guided care vs. SOC.

The feasibility, diagnostic yield, and learning curve of portable echocardiography for out-of-hospital cardiovascular disease screening.

BACKGROUND: The reduction in the size of full-capability echocardiographic machines facilitates "out-of-hospital" transthoracic echocardiography (TTE). Data documenting the feasibility, yield, and logistical considerations of out-of-hospital TTE for preparticipation evaluation of athletes are sparse. METHODS: A multiyear study was conducted to examine the role of 12-lead electrocardiography for athlete screening in which TTE was used to document or exclude underlying structural heart disease. Using a commercially available portable transthoracic echocardiographic system, the rate of technically adequate imaging, diagnostic yield, and the time required for the completion of TTE (including setup, performance, and interpretation) were examined. TTE was performed in university medical offices and at "out-of-office" athletic facilities. Measurements were recorded during each year of the study to determine the impact of targeted attempts to improve efficiency. RESULTS: Four hundred sixty-seven of 510 participants had transthoracic echocardiographic images that were technically adequate for complete interpretation (imaging success rate, 92%). Echocardiographic evidence of physiologic, exercise-induced cardiac remodeling was observed in 110 of 510 (22%). Cardiac abnormalities with relevance to sports participation risk were detected in 11 of 508 participants (2.2%). Over 3 years, the average time for the completion of TTE (including setup, imaging, and interpretation) decreased (year 1, 17.4 ± 3 min; year 2, 14.0 ± 2.1 min; year 3, 11.0 ± 1.8 min; P < .001). This was driven by a significant decrease in the time required for TTE at out-of-office athletic facilities. CONCLUSIONS: Community-based TTE in athletes is feasible and is associated with a high rate of technically adequate imaging. Importantly, there appears to be a significant learning curve associated with out-of-hospital TTE.

Can a teaching intervention reduce interobserver variability in LVEF assessment: a quality control exercise in the echocardiography lab.

OBJECTIVES: This study sought to determine whether a formalized teaching intervention could reduce the interobserver variability (IOV) in visual estimation of left ventricular ejection fraction (LVEF) within a group of sonographers and physicians with a spectrum of experience. BACKGROUND: Precise and reliable echocardiographic assessment of LVEF is necessary for clinical decision-making and minimizing duplicative testing. Skill in the visual estimation of LVEF varies depending on experience and is critical for corroborating EF quantification. IOV may also lead to inconsistency if multiple readers are assessing the EF on serial exams. METHODS: Fourteen cases of 2-dimensional echocardiograms were shown to 25 participants who estimated the EF based on a complete assessment of LV wall motion including parasternal, short-axis, apical, and subcostal views. The cases represented a spectrum of EF range, image quality, and clinical context. Following the initial interpretations, participants underwent a teaching intervention involving tutorial review of reference cases and group discussion of each case with determination of the EF guided by quantitative measure (biplane Simpson method). Three months after the teaching intervention, 14 new cases were shown to the 25 participants following the same methodology. RESULTS: IOV was quantified before and after the teaching intervention with the use of a 3-factor, nested analysis of variance. The factors were: observer, patient, and pre- and post-intervention (time). The analysis of variance showed that the intervention reduced the IOV for the 25 readers between the pre- and post-intervention assessments (F = 2.8, p = 0.007). The IOV decreased from ± 14% EF prior to intervention to ± 8.4% EF following intervention (a 40% reduction in IOV). CONCLUSIONS: In a large echocardiography laboratory with a wide range of training levels and experience, a simple, formalized teaching intervention can successfully diminish IOV of LVEF assessment. This intervention provides not only discrete quality measures, but also serves as a practical tool to document and improve quality of reporting, potentially reducing clinical inefficiencies and repeat testing.

Use of amino-terminal pro-B-type natriuretic peptide to guide outpatient therapy of patients with chronic left ventricular systolic dysfunction.

OBJECTIVES: The aim of this study was to evaluate whether chronic heart failure (HF) therapy guided by concentrations of amino-terminal pro-B-type natriuretic peptide (NT-proBNP) is superior to standard of care (SOC) management. BACKGROUND: It is unclear whether standard HF treatment plus a goal of reducing NT-proBNP concentrations improves outcomes compared with standard management alone. METHODS: In a prospective single-center trial, 151 subjects with HF due to left ventricular (LV) systolic dysfunction were randomized to receive either standard HF care plus a goal to reduce NT-proBNP concentrations ≤1,000 pg/ml or SOC management. The primary endpoint was total cardiovascular events between groups compared using generalized estimating equations. Secondary endpoints included effects of NT-proBNP-guided care on patient quality of life as well as cardiac structure and function, assessed with echocardiography. RESULTS: Through a mean follow-up period of 10 ± 3 months, a significant reduction in the primary endpoint of total cardiovascular events was seen in the NT-proBNP arm compared with SOC (58 events vs. 100 events, p = 0.009; logistic odds for events 0.44, p = 0.02); Kaplan-Meier curves demonstrated significant differences in time to first event, favoring NT-proBNP-guided care (p = 0.03). No age interaction was found, with elderly patients benefitting similarly from NT-proBNP-guided care as younger subjects. Compared with SOC, NT-proBNP-guided patients had greater improvements in quality of life, demonstrated greater relative improvements in LV ejection fraction, and had more significant improvements in both LV end-systolic and -diastolic volume indexes. CONCLUSIONS: In patients with HF due to LV systolic dysfunction, NT-proBNP-guided therapy was superior to SOC, with reduced event rates, improved quality of life, and favorable effects on cardiac remodeling. (Use of NT-proBNP Testing to Guide Heart Failure Therapy in the Outpatient Setting; NCT00351390).

Unusual position of a prosthetic mitral valve.

Extensive calcification of the mitral annulus in patients who require mitral valvereplacement presents a significant challenge to the surgeon. Several techniques, includingdebridement of the calcium, reconstruction of the annulus, and insertion of the prosthesis in a locationother than the annulus, have been used in such patients. We report the echocardiographictechniques used to evaluate the case of a woman with an unusually positioned prosthetic mitralvalve.

Advanced Cardiovascular Sonographer: a proposal of the American Society of Echocardiography Advanced Practice Sonographer Task Force.

Echocardiographic examinations require a well-trained and competent sonographer to obtain proper anatomic and physiologic data to establish an accurate diagnosis for clinical decision-making and patient management. Although the formal education and training of cardiovascular sonographers are evolving, many entry-level and staff sonographers may not have sufficient practical or clinical knowledge of the necessary components of the echocardiographic study for the individual patient's clinical presentation. In many clinical settings, echocardiograms are read after the patient has left the laboratory. Thus, there is a role for a sonographer who can practice at an advanced level in a cardiovascular ultrasound laboratory to ensure a proper echocardiographic examination is performed on every patient. In this setting, an Advanced Cardiovascular Sonographer (ACS) would be able to review the indication for and quality of the examination. If additional images were needed, the ACS would assist the sonographer in obtaining these images, which would lead to the performance of a complete and fully diagnostic examination before the patient had left the echocardiography laboratory. In clinical practice, the quality of the examinations performed would improve, advancements in echocardiographic methods could be taught and incorporated into daily practice, and patients would be better served. The present report is a proposal from the American Society of Echocardiography Advanced Practice Task Force that identifies the potential of cardiac sonographers to achieve the ACS level.

Persistent and reversible cardiac dysfunction among amateur marathon runners.

AIMS: Transient systolic and diastolic abnormalities in ventricular function have previously been documented during endurance sports. However, these described alterations may be limited by the techniques applied. We sought, using less load-dependent methods, to characterize both the extent and the chronology of the cardiac changes associated with endurance events. METHODS AND RESULTS: Transthoracic echocardiography (TTE) was performed prior to, immediately after, and approximately 1 month after completion of the 2003 Boston Marathon in 20 amateur athletes. TTE included two-dimensional, spectral and tissue Doppler (TD) and flow propagation velocity (V(p)). After completion of the marathon, global measures of left ventricular (LV) systolic function were unchanged (EF 59 +/- 6 vs. 61 +/- 4% post, P = 0.14), whereas TD-derived measures of LV systolic function [septal strain -23 +/- 5 vs. -17 +/- 4%, P = 0.007; septal strain rate (SR) -1.5 +/- 0.3 vs. -1.1+/- 0.2 s(-1), P = 0.007] and right ventricular (RV) systolic function (RV apical strain -33 +/- 4 vs. -27 +/- 5%, P = 0.001; RV apical SR -2.4 +/- 0.7 vs. -1.8 +/- 0.5, P = 0.002) were reduced. Significant changes in transmitral velocity (E/A ratio 2.0 +/- 0.5 vs.1.3 +/- 0.3, P = 0.005) and TD indices of LV and RV diastolic function (E(a) septal 9.5 +/- 1.8 vs. 8.1 +/- 1.2 cm/s post-marathon, P = 0.01) were also observed, indicating an inherent alteration in LV relaxation. Although all indices of LV and RV systolic function had returned to normal on follow-up, there were persistent diastolic abnormalities (RV E(a), 11.5 +/- 1.5 cm/s pre-marathon vs. 10.0 +/- 1.6 cm/s follow-up, P = 0.01). CONCLUSION: Marathon running leads to transient systolic and more persistent diastolic dysfunction of both the LV and the RV.

Myocardial adaptation to short-term high-intensity exercise in highly trained athletes.

We aimed to clarify the myocardial adaptation to short-term high-intensity exercise among trained athletes. We screened 17 participants in the 2004 World Indoor Rowing Championships before and after a 2000-m sprint. Echocardiography included standard measurements and tissue Doppler-derived strain (epsilon), strain rate, and 2-dimensionally derived speckle-tracking imaging for left ventricular (LV) torsion. LV volumes and ejection fraction were unchanged after exercise. There was a reduction in early and an increase in late diastolic filling velocities and a decrease in the flow propagation velocity. Annular systolic velocities, slope of the systolic acceleration, septal and lateral epsilon, and speckle tracking-derived torsion were increased. The increased LV torsion was a result of increased basal and apical rotation. Right ventricular apical epsilon decreased. In conclusion, maximal intensity short-duration exercise was associated with attenuation of LV diastolic function, augmentation of LV systolic function, and a reduction in apical right ventricular contractility.

A locus for autosomal dominant mitral valve prolapse on chromosome 11p15.4.

Mitral valve prolapse (MVP) is a common cardiovascular abnormality in the United States, occurring in approximately 2.4% of the general population. Clinically, patients with MVP exhibit fibromyxomatous changes in one or both of the mitral leaflets that result in superior displacement of the leaflets into the left atrium. Although often clinically benign, MVP can be associated with important accompanying sequelae, including mitral regurgitation, bacterial endocarditis, congestive heart failure, atrial fibrillation, and even sudden death. MVP is genetically heterogeneous and is inherited as an autosomal dominant trait that exhibits both sex- and age-dependent penetrance. In this report, we describe the results of a genome scan and show that a locus for MVP maps to chromosome 11p15.4. Multipoint parametric analysis performed by use of GENEHUNTER gave a maximum LOD score of 3.12 for the chromosomal region immediately surrounding the four-marker haplotype D11S4124-D11S2349-D11S1338-D11S1323, and multipoint nonparametric analysis (NPL) confirms this finding (NPL=38.59; P=.000397). Haplotype analysis across this region defines a 4.3-cM region between the markers D11S1923 and D11S1331 as the location of a new MVP locus, MMVP2, and confirms the genetic heterogeneity of this disorder. The discovery of genes involved in the pathogenesis of this common disease is crucial to understanding the marked variability in disease expression and mortality seen in MVP.