Echocardiographic Image Quality Research Articles

A multi-task deep learning approach for real-time view classification and quality assessment of echocardiographic images

High-quality standard views in two-dimensional echocardiography are essential for accurate cardiovascular disease diagnosis and treatment decisions. However, the quality of echocardiographic images is highly dependent on the practitioner’s experience. Ensuring timely quality control of echocardiographic images in the clinical setting remains a significant challenge. In this study, we aimed to propose new quality assessment criteria and develop a multi-task deep learning model for real-time multi-view classification and image quality assessment (six standard views and “others”). A total of 170,311 echocardiographic images collected between 2015 and 2022 were utilized to develop and evaluate the model. On the test set, the model achieved an overall classification accuracy of 97.8% (95%CI 97.7–98.0) and a mean absolute error of 6.54 (95%CI 6.43–6.66). A single-frame inference time of 2.8 ms was achieved, meeting real-time requirements. We also analyzed pre-stored images from three distinct groups of echocardiographers (junior, senior, and expert) to evaluate the clinical feasibility of the model. Our multi-task model can provide objective, reproducible, and clinically significant view quality assessment results for echocardiographic images, potentially optimizing the clinical image acquisition process and improving AI-assisted diagnosis accuracy.

Echocardiographic image quality deteriorates with the severity of cardiogenic shock

BackgroundTransthoracic echocardiography (TTE) is the primary tool for assessing left ventricular (LV) function in cardiogenic shock (CS). However, inadequate image quality often hinders it. In this retrospective study, we investigated factors associated with LV image quality in patients admitted to the intensive care unit (ICU) with ischemic CS.ResultsTwo critical care physicians accredited in echocardiography independently reviewed the TTEs of 100 patients admitted to our tertiary cardiac ICU with ST-elevation myocardial infarction complicated by CS between October 2016 and September 2019. Endocardial border definition (EBD) was graded for each myocardial segment of the apical 4-chamber and 2-chamber views using a conventional scoring system (1 = good, 2 = suboptimal, 3 = poor, and 4 = not possible). The biplane EBD index (EBDi) was calculated by averaging all segments from both views. An average EBDi of both observers was correlated with clinical and echocardiographic parameters. The median age was 62 years [54, 73], and 78% were males. LV ejection fraction and cardiac index (CI) medians were 29% [20, 35] and 1.93 l/min/m2 [1.40, 2.51], respectively. The median biplane EBDi was nearly suboptimal (1.833 [1.542, 2.083]). There was no correlation between EBDi and age, sex, or body mass index. However, biplane EBDi demonstrated statistically significant correlations with PaO2 (r2 = 0.066, p = 0.01), mean arterial pressure (MAP, r2 = 0.055, p = 0.03), CI (r2 = 0.105, p < 0.01), tricuspid annulus systolic velocity (RV S’, r2 = 0.092, p = 0.01), and tricuspid regurge maximum velocity (TR Vmax, r2 = 0.067, p = 0.01). In a multivariate model, only CI correlated independently with EBDi (r2 = 0.105, p < 0.01). The biplane EBDi predicted CI (area under the curve (AUC) 0.70, p = 0.001) with good sensitivity (71%) and reasonable specificity (61%).ConclusionsThe study suggests that in patients admitted to the ICU with ischemic CS, LV image quality by TTE deteriorates with the severity of shock, as indicated by CI.

Biomedical Applications of Wavelet Transform Algorithm on Deep Learning Ultrasonic Image Optimization as a Prognosis Model for Acute Myocarditis

We aimed to investigate the biomedical methods of wavelet transform algorithm on ultrasonic image denoising algorithm and the risk factors for the adverse prognostic events of patients with myocarditis and analyse its correlation with free triiodothyronine (FT3) level. A retrospective study was performed to include 68 patients diagnosed with acute myocarditis (AM). The included patients were enrolled into adverse event (AE) group (n = 7) and non-adverse event (NAE) group (n = 61). The clinical data, laboratory examination indicators, echocardiographic parameters, and thyroid functions between the patients in the two groups at admission were compared. Besides, wavelet transform (WT) algorithm was employed to process ultrasonic images containing noises. Univariate and multivariate analysis were performed using Logistic regression model. It was demonstrated that peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) (35.279, 0.847) of wavelet transform algorithm were higher than those of denoising convolutional neural network (DnCNN) algorithm (30.673, 0.582) and Red-Net algorithm (28.489, 0.638). In the AE group, the QRS interval was longer ([102 (93, 135) ms] versus [86 (79, 102) ms]), the proportion of prolonged QRS period was higher (28.57%, 9.83%), and the creatine kinase isoenzyme, a marker of myocardial injury, was higher [32.87 (13.64, 78.62) U/L] versus 12.02 (9.89, 27.65) U/L], and the level of FT3 was lower [1.87 (1.23, 2.36) pg/mL versus 2.83 (1.83, 3.45) pg/mL] compared to the NAE group. The left ventricular ejection fraction (LVEF) in the adverse event group was lower than that in the non-adverse event group [45.78 (36.18, 54.32) % versus 63.72 (54.82, 64.68)]. Multivariate Logistic regression analysis showed that the risk factors for adverse events in patients with acute myocarditis included QRS interval &gt; 120 ms (OR = 1.021), creatine kinase isoenzyme &gt; 24 U/L (OR = 1.024), FT3 (OR = 0.067), and LVEF &lt; 50% (OR = 0.973). This work confirmed that the wavelet transform algorithm can optimize the image quality of echocardiography, improve the clarity, and provide a feasible idea for improving the prognosis of patients with acute myocarditis.

Is myocardial fibrosis in patients with non-ischemic dilated cardiomyopathy associated with the whole-heart myocardial mechanics?

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Non-ischemic dilated cardiomyopathy (NIDCM) is a severe form of primary myocardial disease related with progressive heart failure (HF) and a poor prognosis. In patients with NIDCM, myocardial fibrosis (MF) is associated to adverse cardiovascular outcomes and should be evaluated quickly in each case of NIDCM [1,2]. Purpose To evaluate the relationships between the whole-heart myocardial strain parameters and presence of late gadolinium enhancement (LGE) in a prospective cohort of NIDCM patients. Methods The study group consisted of 101 patients with NIDCM who were referred for a cardiovascular magnetic resonance (CMR) and speckle tracking echocardiography examination. We analyzed MF by LGE, and the whole-heart (both ventricles and atria) myocardial mechanics were evaluated by speckle tracking. All patients underwent a comprehensive clinical evaluation, laboratory test assessment, electrocardiogram, Holter monitoring. The exclusion criteria were ischemic coronary disease, primary valvular heart disease, chronic kidney disease (eGFR &amp;lt;30 ml/min/1.73 m2), under the age of 18, a poor echocardiographic or CMR image quality, inflammatory myocardial disease, previous pulmonary embolism, peripartum cardiomyopathy and toxic damage. Results The study group was divided into two groups according to the presence of MF fibrosis. The presence of MF was associated to worse of the whole-heart myocardial strain parameters. Both ventricles or atria were more dilated in the group with MF. The strongest relationships were detected between presence of MF and mechanical strain parameters of the left side of the heart (left ventricle ejection fraction (LVEF) (r=0.662, 25 p&amp;lt;0.001), left ventricle global circumferential strain (LV GCS) (r=-0.609, p&amp;lt;0.001), left ventricle global longitudinal strain (LV GLS) (r=-0.586, p&amp;lt;0.001), left atrial reservoir strain (LASr) (r=0.588, p&amp;lt;0.001). Logistic regression analysis revealed that the last three mentioned strain parameters were independently associated with presence of MF. Conclusions In patients with NIDCM, the whole-heart myocardial mechanics are worse in patients‘ presence of MF. LV GLS, LV GCS, and LARs are the best 2D echocardiographic mechanical strain parameters to prognosticate the presence of MF.

Left ventricular longitudinal strain: a useful tool in the evaluation of patients with Takotsubo syndrome

Abstract Funding Acknowledgements Type of funding sources: None. Introduction In contrast with the usual type of Takotsubo syndrome (TS), where apical wall motion abnormality occurs, the less common patterns (mid-ventricular, basal, and focal) are often misdiagnosed. Purpose In this retrospective study we examined the role of global (GLS) and territorial longitudinal strain with speckle tracking in the evaluation of patients with TS. Methods Patients with TS were examined at the echocardiography laboratory of our general hospital, from August 2019 until August 2022. Left ventricular ejection fraction (LVEF) and GLS were calculated and territorial longitudinal strain was assessed. Results Out of 21 patients with TS (19 females), of median age 67.5 years (range 33.0 – 87.0 years), 11 (52.4%) had apical distribution, 6 (28.6%) mid-ventricular, 2 (9.5%) basal and 2 (9.5%) focal. Of those, 3 were excluded from the study due to insufficient echocardiographic image quality. Patients with apical pattern had median LVEF 37% (range 23–51%) and GLS −7.6% (−5.8 to −12.5%), patients with mid-ventricular pattern had median LVEF 40% (35–50%) and GLS −14.2% (−9.3 to −16.5%) and patients with basal pattern had median LVEF 52% (47–56%) and GLS −16.5% (−14.6 to −18.4%). In the first case of focal distribution, wall motion abnormality was found at the anterior septal wall, and LVEF was calculated 46% and GLS −12.8%. In the second case, abnormality was found at the inferoposterior wall and LVEF was calculated 56% and GLS −17.3%. Patients with apical TS had median strain of basal territory −11.7%, mid territory −3.7% and apical territory +3.6%. Patients with mid ventricular TS had median strain of basal territory −10.2%, mid territory −6.5% and apical territory −19.7%. Patients with basal TS had median strain of basal territory −4.3%, mid territory −16.4% and apical territory −29.1%. Conclusion GLS and territorial longitudinal strain reflect the reduction in left ventricular systolic function and the distribution of wall motion abnormalities in patients with TS, which further supports the initial diagnosis.

Machine learning derived echocardiographic image quality in patients with left ventricular systolic dysfunction: insights on the echo views of greatest image quality.

We sought to determine the cardiac ultrasound view of greatest quality using a machine learning (ML) approach on a cohort of transthoracic echocardiograms (TTE) with abnormal left ventricular (LV) systolic function. We utilize an ML model to determine the TTE view of highest quality when scanned by sonographers. A random sample of TTEs with reported LV dysfunction from 09/25/2017-01/15/2019 were downloaded from the regional database. Component video files were analyzed using ML models that jointly classified view and image quality. The model consisted of convolutional layers for extracting spatial features and Long Short-term Memory units to temporally aggregate the frame-wise spatial embeddings. We report the view-specific quality scores for each TTE. Pair-wise comparisons amongst views were performed with Wilcoxon signed-rank test. Of 1,145 TTEs analyzed by the ML model, 74.5% were from males and mean LV ejection fraction was 43.1 ± 9.9%. Maximum quality score was best for the apical 4 chamber (AP4) view (70.6 ± 13.9%, p<0.001 compared to all other views) and worst for the apical 2 chamber (AP2) view (60.4 ± 15.4%, p<0.001 for all views except parasternal short-axis view at mitral/papillary muscle level, PSAX M/PM). In TTEs scanned by professional sonographers, the view with greatest ML-derived quality was the AP4 view.

An automated method for BRISQUE quantification of image quality in echocardiography

Abstract Background Echocardiography (echo) remains the most widely used imaging modality for the assessment, monitoring, and prognostication of the heart. Despite its prevalence, standardisation efforts for echo chamber quantification are ongoing, with challenges owing to subjectivity during acquisition and analysis. Furthermore, the confidence in derived functional indices is often dependent on the quality of the acquired images. However, few studies have investigated the accuracy of echo measurements compared to a reference modality such as cardiac magnetic resonance (CMR) imaging, when stratified by image quality. Purpose To develop an objective and automated method to quantify echo image quality, and subsequently to investigate the relationship between image quality and patient demographics, as well as the magnitude of bias in left ventricular (LV) functional indices compared with CMR. Methods Transthoracic apical 2D echo (2DE) and 3D echo (3DE) data from 128 participants (72 healthy controls and 56 patients with acquired heart disease) were used to train a BRISQUE (Blind/Referenceless Image Spatial Quality Evaluator) algorithm [1]. Briefly, feature extraction was performed by fitting pixel luminances to a generalised Gaussian distribution (GGD), followed by support vector regression to correlate features (i.e., shape, variance, and mean parameters of the GGD) to quality scores (Fig. 1). Independent BRISQUE models were trained on 580 2DE images (consisting of 2-, 3-, and 4-chamber views) and 128 targeted LV 3DE acquisitions at end-diastole, each assigned a subjective perceived quality score between 1 (poor) and 9 (excellent) by a single observer. LV indices including end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), and global longitudinal strain (GLS), were assessed according to standard guidelines. Resultant BRISQUE scores were plotted against patient demographics (age, height, weight) and the measurement bias by comparison to CMR (acquired within 1 hour of echo). Results Several linear relationships (where P-value of slope &amp;lt;0.05) were observed between demographics, cardiac indices, and BRISQUE scores. Increasing patient weight (and height in 3DE) were found to be associated with poorer image quality. There was no apparent relationship between image quality and age. Of interest, EF exhibited a relationship with image quality in both 2DE and 3DE (Fig. 2), whereby higher quality images tended to overestimate EF, while lower quality images underestimated EF. For 3DE, image quality dependency was also observed for ESV and GLS biases. Conclusions BRISQUE can objectively quantify image quality to produce scores which correlate to those of an expert observer, with potential utility for the standardised quantification of echo image quality. Using this method, it may be possible to predict patient characteristics which adversely impact echo quality, as well as the magnitude of measurement biases for certain functional indices. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Health Research Council (HRC) of New Zealand; National Heart Foundation (NHF) of New Zealand

An automated method for BRISQUE quantification of image quality in echocardiography

Abstract Background Echocardiography (echo) remains the most widely used imaging modality for the assessment, monitoring, and prognostication of the heart. Despite its prevalence, standardisation efforts for echo chamber quantification are ongoing, with challenges owing to subjectivity during acquisition and analysis. Furthermore, the confidence in derived functional indices is often dependent on the quality of the acquired images. However, few studies have investigated the accuracy of echo measurements compared to a reference modality such as cardiac magnetic resonance (CMR) imaging, when stratified by image quality. Purpose To develop an objective and automated method to quantify echo image quality, and subsequently to investigate the relationship between image quality and patient demographics, as well as the magnitude of bias in left ventricular (LV) functional indices compared with CMR. Methods Transthoracic apical 2D echo (2DE) and 3D echo (3DE) data from 128 participants (72 healthy controls and 56 patients with acquired heart disease) were used to train a BRISQUE (Blind/Referenceless Image Spatial Quality Evaluator) algorithm [1]. Briefly, feature extraction was performed by fitting pixel luminances to a generalised Gaussian distribution (GGD), followed by support vector regression to correlate features (i.e., shape, variance, and mean parameters of the GGD) to quality scores (Fig. 1). Independent BRISQUE models were trained on 580 2DE images (consisting of 2-, 3-, and 4-chamber views) and 128 targeted LV 3DE acquisitions at end-diastole, each assigned a subjective perceived quality score between 1 (poor) and 9 (excellent) by a single observer. LV indices including end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), and global longitudinal strain (GLS), were assessed according to standard guidelines. Resultant BRISQUE scores were plotted against patient demographics (age, height, weight) and the measurement bias by comparison to CMR (acquired within 1 hour of echo). Results Several linear relationships (where P-value of slope &amp;lt;0.05) were observed between demographics, cardiac indices, and BRISQUE scores. Increasing patient weight (and height in 3DE) were found to be associated with poorer image quality. There was no apparent relationship between image quality and age. Of interest, EF exhibited a relationship with image quality in both 2DE and 3DE (Fig. 2), whereby higher quality images tended to overestimate EF, while lower quality images underestimated EF. For 3DE, image quality dependency was also observed for ESV and GLS biases. Conclusions BRISQUE can objectively quantify image quality to produce scores which correlate to those of an expert observer, with potential utility for the standardised quantification of echo image quality. Using this method, it may be possible to predict patient characteristics which adversely impact echo quality, as well as the magnitude of measurement biases for certain functional indices. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Health Research Council (HRC) of New Zealand; National Heart Foundation (NHF) of New Zealand

Cardiooncology-usefulness of cardiac MRI : Inflammation, fibrosis, outcome

With rapidly increasing survival chances of cancer patients, the potential side effects of cancer therapeutics are increasingly relevant and apotentially lifelong issue. If cardiotoxic effects are not detected at areversible stage, this might result in irreversible heart failure. This article will portray the current state of knowledge on the use of cardiac magnetic resonance imaging (cardiac MRI) in the field of cardio-oncology. The aim is to provide an overview of the advantages of cardiac MRI to determine myocardial function and analyze inflammatory or fibrotic myocardial changes. Current studies on this topic were collected and evaluated. Expert recommendations from various position papers were reviewed and summarized. Lastly, anMRI protocol to assess potential cardiotoxic effects of cancer therapeutics was discussed. Up to 20% of patients are reported to suffer from cancer therapeutics-related cardiac dysfunction (CTRCD). Especially those with cardiovascular risk factors should receive pre- and posttherapeutic monitoring of heart function. Cardiac MRI is currently suggested as an imaging tool to analyze myocardial function if echocardiographic assessment is insufficient. However, cardiac MRI is also an excellent method for additional tissue analysis. Current consensus statements recommend cardiac MRI as optional in cases where echocardiography image quality is not adequate. Nevertheless, patients with reduced heart function on echocardiography might benefit from early assessment of inflammatory or fibrotic changes due to CTRCD using cardiac MRI.

P108 PLEURAL EFFUSION AND THE POSTERIOR THORACIC WINDOW: ONE MORE CHANCE TO IMAGE THE AORTIC VALVE

Abstract A 85–year–old woman was admitted to our institution with acute respiratory distress. Clinical examination revealed a 5/6 systo–diastolic murmur radiating to both carotid arteries and bilateral lung crackles. Laboratory workup was notable for a N–terminal pro–brain natriuretic peptide (NT–proBNP) of 15.000 ng/L (normal range: &amp;lt; 300 ng/L) and a moderate high–sensitive troponin T elevation of 245 ng/L (normal range: &amp;lt; 14 ng/L) with no increase on serial testing. The hemoglobin and renal function was normal. Evaluation by chest X–ray showed bilateral diffuse pulmonary infiltrates consistent with pulmonary edema and bilateral pleural effusion. The patient was admitted in CCU. A transthoracic echocardiogram was performed bedside, but it was extremely difficult to obtain good quality images from standard view given the ortopnoic decubitus of the patient. However, the aortic valve was heavily calcified and severe aortic stenosis was documented with a mean gradient of 42 mmHg from apical 5 chambers view; 35 mmHg from right parasternal view, with an functional aortic valve area ≈0.6 cm2 (figure, upper panel). Left ventricle shows eccentric hypertrophy and a moderate diffuse reduction of ejection fraction (EF 35–40%). Then, the patient was positioned sitting upright, because the left pleural effusion offered an additional acoustic window—the posterior thoracic window (PTW)—that allowed better alignment of the ultrasound beam with the aortic jet. Mean gradient was recorded definitely higher than from standard view at 50 mm Hg. Moreover, aortic regurgitation that appeared mild from standard view was documented as significant from PTW. In fact, a proper alignment of the aortic regurgitation jet was feasible from this view with a quantification of severity with multiparameter criteria as suggested from the guidelines (PHT 166 ms, Vena contracta 8 mm; Rvol 60 ml). (Figure, lower panel) Conclusion Obtaining adequate echocardiographic images in critically ill patients is important for better diagnosis and treatment. For several reasons, this group of patients remains among the most challenging with regard to quality of echocardiographic images. In presence of pleural effusion and technically difficult echo exams, the PTW should be considered in the assessment of cardiac structures including aortic valve as a potentially useful option to provide further diagnostic information.

Echocardiography image quality of global longitudinal strain in cardio-oncology: a prospective real-world investigation.

Left-ventricular (LV) global longitudinal strain (GLS) has been reported to be a robust and sensitive marker of chemotherapy-induced cardiac damage. Image quality is paramount for accurate GLS measurements. In real-world cardio-oncology settings, the incidence of suboptimal echocardiography quality and its significance in clinical decision-making have not been well investigated. This prospective study examined the incidence and impact of suboptimal echocardiographic image quality on detecting subtle myocardial damage by chemotherapy. Seventy-seven consecutive patients with breast cancer (age, 52 ± 12years, 76 women, 33 with left-sided breast cancer) were included in this study. Echocardiography was performed at 3-month intervals 1year before and after chemotherapy initiation. We classified the image quality of each echocardiographic acquisition into three groups: optimal, suboptimal, or inadequate for speckle tracking. Among the 376 examinations obtained during the cardiac monitoring, the image quality in 194 (52%) was optimal, suboptimal in 159 (42%), and inadequate in 23 (6%). The interobserver reproducibility was 0.91 in the optimal and 0.21 in the suboptimal group. In contrast, the optimal group showed progressive impairment in both GLS (p = 0.001) and LV ejection fraction (LVEF) (p < 0.001) during follow-up, and the suboptimal group showed a progressive decrease in LVEF (p = 0.006), but not in GLS (p = 0.13). Left-sided mammotomy and/or reconstruction surgery and high body mass index were significant determinants of suboptimal image quality. Even in cases of minor image quality impairment, the physician should assess GLS carefully to avoid errors in crucial clinical decision-making.

Color kinesis-dobutamine stress echocardiography pinpoints coronary artery disease

Background/Aim: Dobutamine stress echocardiography (DSE) can identify significant coronary artery disease (CAD) and where it is localized. However, determining endocardial borders with poor echocardiographic views may create unsatisfactory results. Color kinesis (CK) shows endocardial movement with color, and allows easier and more objective evaluation of ventricular wall motion. In this study, our aim was to evaluate the role of CK in determining CAD localization during DSE. Methods: The study group consists of patients whose CAD diagnosis was confirmed with coronary angiography (CA). Patients with atrial fibrillation (A-Fib), left bundle branch block, poor echocardiography image quality, left ventricular (LV) ejection fraction < 40%, and non-ischemic LV wall motion abnormality were excluded. CK-DSE and dobutamine stress-induced myocardial perfusion scintigraphy (MPS) was done in all patients and compared to CA. Results: A total of twenty patients [16 males (80%) and 4 females (20%)] were included in the study. CK-DSE results were consistent with CA in determining CAD localization (kappa 0.66). Vessel-based kappa values for LAD, RCA, and Cx were 0.81, 0.70 and 0.61, respectively. Consistency between MPS and CK-DSE was evaluated by the area under curves (AUC) within a ROC curve analysis (p > 0.05, 95% CI). Conclusion: Our study showed that CK allows for rapid, objective, and automatic evaluation of segmental wall motion. In addition, CK-DSE is consistent with CA results.

Full cardiac cycle asynchronous temporal compounding of 3D echocardiography images.

It is important to improve echocardiography image quality, because the accuracy of echocardiographic assessment and diagnosis relies on image quality. Previous work on 2D temporal image compounding for image frames with matching cardiac phases (synchronous), and for temporally neighbouring image frames (asynchronous) over small ranges of time frames showed good improvement to image quality. Here, we extend this by performing asynchronous temporal compounding to echocardiographic images in 3D, involving all frames within a cardiac cycle, via a robust 3D cardiac motion estimation algorithm to describe the large image deformations. After compounding, the images can be reanimated via the motion model. Various methods of fusing image frames together are tested, including mean, max, and wavelet methods, and outlier rejection algorithms. The compounding algorithm is applied on 3D human adult, porcine adolescent, and human fetal echocardiography images. Results show significant improvements to contrast-to-noise ratio (CNR) and boundary clarity, and significantly decreased variability in manual quantification of cardiac chamber volumes after compounding. Interestingly, compounding can extend the field of view of the echo images, by reconstructing cardiac structures that momentarily exceeded the field of view, using the motion estimation algorithm to calculate their locations outside the field of view during these time periods. Although all compounding methods provide general improvements, the mean method led to blurred boundaries, while the max methods led to high variability of CNR. Outlier rejection algorithms were found to be useful in addressing these weaknesses.

The Effect of Obesity on Echocardiographic Image Quality

Increased body mass index (BMI) may reduce transthoracic echocardiogram (TTE) image quality, resulting in increased requirements for ultrasound enhancing agents (UEA), as recommended by the American Society of Echocardiography (ASE), and a greater incidence of non-diagnostic studies. Over a 5-month period 1,108 TTEs were analysed as to (1) whether they could answer the clinical question posed by the ordering physician (i.e. were diagnostic vs non diagnostic), and (2) whether they required UEAs according to the ASE guidelines. Patient characteristics were gathered from the medical record. 12.9% of TTEs were non-diagnostic (21.0% of TTEs in the obese population [BMI≥30 kg/m2] vs 7.8% in the non-obese [p<0.001]). Predictors of a non-diagnostic study were BMI (OR 1.09, [95% CI 1.06-1.11], p<0.0001), male gender (OR 1.54, [1.06-2.25], p=0.02), and inpatient status (OR 1.75, [1.20-2.55], p=0.004). Obesity (BMI≥30) was strongly associated with non-diagnostic studies (OR 3.22, [2.23-4.51], p<0.001). Factors associated with increased requirement of UEAs were BMI (OR 1.10, [1.08-1.12], p<0.0001), age (OR 1.02, [1.01-1.03], p<0.0001) and inpatient status (OR 1.7, [1.29-2.24], p<0.05). Obesity (BMI>30) was strongly associated with contrast requirement (OR 3.16, [2.43-4.10], p<0.0001). Body mass index, male gender and inpatient status were associated with an increased incidence of non-diagnostic studies. Body mass index, age and inpatient status were associated with an increased requirement for UEAs.

The Role of Four-Dimensional Automatic Right Ventricular Quantification Technology to Determine RV Function and Hemodynamics in Patients With Pulmonary Hypertension Compared With Right Heart Catheterization.

Background: Four-dimensional automatic right ventricular quantification technology (4D auto-RVQ) is a new method that can simultaneously measure right ventricular (RV) structure and strain. The role of 4D auto-RVQ in determining RV function and hemodynamics is not clear. The role of 4D auto-RVQ in determining RV function and hemodynamics is not clear. We assessed the 4D auto-RVQ to measure right heart structure, function, and hemodynamics in patients with pulmonary hypertension (PHTN) correlated with right heart catheterization (RHC).Methods: We enrolled a prospective cohort of 103 patients with PHTN and 25 healthy controls between September 2017 and December 2018. All patients with PHTN underwent echocardiography and RHC. Patients were included if they underwent two-dimensional (2D) and 4D auto-RVQ echocardiographic sequences on the same day as RHC. We analyzed RV functional indices using 2D and 4D auto-RVQ analyses. We divided patients with PHTN into three groups according to echocardiographic image quality as follows: high (n = 24), average (n = 48), and poor (n = 4). Hemodynamic parameters were measured using RHC, including mean right atrial pressure, mean pulmonary arterial pressure, RV cardiac index (RV-CI), and pulmonary vascular resistance.Results: There were significant differences in most 2D and 4D auto-RVQ parameters between patients with PHTN and healthy controls. Interobserver variability showed significant agreement with 4D auto-RVQ for most measurements except for 4D end-diastolic volume. Indices measured by auto 4D-RVQ in the high-quality image group had a good correlation with RHC but not in the average- and poor-quality image group. Mid-RV diameter showed the best predictive power for the right RV-CI [area under the curve (AUC) 0.935; 95% confidence interval (CI), 0.714–0.997; p < 0.001]. RV end-systolic volume >121.50 mL had a 71.43% sensitivity and a 100% specificity to predict right RV-CI (AUC, 0.890; 95% CI, 0.654–0.986; p < 0.001).Conclusions: 4D auto-RVQ may be used to estimate RV function and some hemodynamic changes compared with RHC in PHTN patients with high image quality. Furthermore, a large sample of the study is needed to evaluate RV function by 4D auto-RVQ in PHTN patients with average image quality.

Maximal Wall Thickness Measurement in Hypertrophic Cardiomyopathy: Biomarker Variability and its Impact on Clinical Care

ObjectivesThe aim of this study was to define the variability of maximal wall thickness (MWT) measurements across modalities and predict its impact on care in patients with hypertrophic cardiomyopathy (HCM). BackgroundLeft ventricular MWT measured by echocardiography or cardiovascular magnetic resonance (CMR) contributes to the diagnosis of HCM, stratifies risk, and guides key decisions, including whether to place an implantable cardioverter-defibrillator (ICD). MethodsA 20-center global network provided paired echocardiographic and CMR data sets from patients with HCM, from which 17 paired data sets of the highest quality were selected. These were presented as 7 randomly ordered pairs (at 6 cardiac conferences) to experienced readers who report HCM imaging in their daily practice, and their MWT caliper measurements were captured. The impact of measurement variability on ICD insertion decisions was estimated in 769 separately recruited multicenter patients with HCM using the European Society of Cardiology algorithm for 5-year risk for sudden cardiac death. ResultsMWT analysis was completed by 70 readers (from 6 continents; 91% with >5 years’ experience). Seventy-nine percent and 68% scored echocardiographic and CMR image quality as excellent. For both modalities (echocardiographic and then CMR results), intramodality inter-reader MWT percentage variability was large (range –59% to 117% [SD ±20%] and –61% to 52% [SD ±11%], respectively). Agreement between modalities was low (SE of measurement 4.8 mm; 95% CI 4.3 mm-5.2 mm; r = 0.56 [modest correlation]). In the multicenter HCM cohort, this estimated echocardiographic MWT percentage variability (±20%) applied to the European Society of Cardiology algorithm reclassified risk in 19.5% of patients, which would have led to inappropriate ICD decision making in 1 in 7 patients with HCM (8.7% would have had ICD placement recommended despite potential low risk, and 6.8% would not have had ICD placement recommended despite intermediate or high risk). ConclusionsUsing the best available images and experienced readers, MWT as a biomarker in HCM has a high degree of inter-reader variability and should be applied with caution as part of decision making for ICD insertion. Better standardization efforts in HCM recommendations by current governing societies are needed to improve clinical decision making in patients with HCM.

Reliability of left atrial strain reference values: A 3D echocardiographic study.

Standard apical four-chamber and two-chamber views often maximize the long-axis of the left ventricle, resulting in artifactitious foreshortening of the left atrium (LA), which may overestimate LA longitudinal reservoir strain (LALS). We compared LALS values between 2D echocardiography (2DE) and 3D echocardiography (3DE) in healthy subjects to determine whether 2DE speckle tracking analysis overestimates the reference value of LALS. In this study, 4 types of cohorts were included: 1. 105 normal subjects (retrospectively), 2. 53 patients with cardiovascular diseases (retrospectively), 3. 15 patients who received cardiac magnetic resonance (prospectively), and 4. 20 normal subjects (prospectively). LALS and LA length were measured using both 2DE and 3DE in 105 healthy subjects (median age: 42 years). Biplane LALS was measured in apical four- and two-chamber views using 2DE speckle tracking software, and 3DE LALS was measured using new 3DE LA strain software. To determine sensitivity, we also performed the same analysis in 53 patients with cardiovascular disease. The mean value of biplane LALS was 39.6%. LA length at both end-diastole (r = -0.43) and end-systole (r = -0.54) was negatively correlated with biplane LALS. Multivariate regression analysis revealed that both end-diastolic and end-systolic LA length had significant negative relationships with biplane LALS after adjusting for anthropometric and echocardiographic image quality parameters. 3DE LALS (23.7±7.6%) gave significantly lower values than 2DE LALS (39.5±12.0%, p<0.001) with a weak correlation (r = 0.33). LA length measured by 2DE was significantly shorter than that measured by 3DE. The same trend was observed in diseased patients. Our results revealed that in 2DE, the LA cavity consistently appears longitudinally foreshortened in apical views, potentially overestimating LALS. 3DE may overcome this limitation.

Impact of Reverse Left Ventricular Remodeling on Outcomes of Patients with Anomalous Left Coronary Artery from the Pulmonary Artery after Surgical Correction.

The impact of reverse left ventricular remodeling (r-LVR) on clinical outcomes after surgical correction of anomalous left coronary artery from the pulmonary artery (ALCAPA) remains unclear. This study aims to examine the prognostic significance of r-LVR in patients with ALCAPA after surgery. We prospectively identified 61 patients undergoing surgical correction for ALCAPA; 54 patients had adequate echocardiographic image quality with quantitative biplane analysis performed both at baseline and at 30-day postoperative follow-up. Postoperative r-LVR was defined as a reduction of ≥ 10% in left ventricular end-diastolic volume index during follow-up. Cox proportional-hazards regression was used to investigate the independent association of r-LVR and all-cause mortality. Among 54 patients (age: 21.2 ± 7 months; 37% females), r-LVR occurred in 35 patients (64.8%) after surgery. Compared to patients with r-LVR, patients without r-LVR had significantly higher level of N-terminal pro B-type natriuretic peptide (NT-proBNP) [2176 (711, 4219) vs 998 (623, 2145) P < 0.001] and lower survival rate (47.3% vs 82.9%, HR = 5.72 [1.96 to 17.20], P < 0.001) at 1-year follow-up. NT-proBNP (OR = 2.27 [1.67 to 18.3], P = 0.02) was an independent predictor of r-LVR in multivariate analysis. Moreover, r-LVR was significantly associated with a lower rate of all-cause mortality (HR = 0.27 [0.08 to 0.98], P = 0.03) in multivariate analysis, even after adjustment for clinical and echocardiographic variables. R-LVR occurred in more than half of patients with ALCAPA undergoing surgical correction and it was associated with better clinical outcomes. NT-proBNP is an independent predictor of r-LVR.

Automated estimation of echocardiogram image quality in hospitalized patients.

We developed a machine learning model for efficient analysis of echocardiographic image quality in hospitalized patients.This study applied a machine learning model for automated transthoracic echo (TTE) image quality scoring in three inpatient groups. Our objectives were: (1) Assess the feasibility of a machine learning model for echo image quality analysis, (2) Establish the comprehensiveness of real-world TTE reporting by clinical group, and (3) Determine the relationship between machine learning image quality and comprehensiveness of TTE reporting. A machine learning model was developed and applied to TTEs from three matched cohorts for image quality of nine standard views. Case TTEs were comprehensive studies in mechanically ventilated patients between 01/01/2010 and 12/31/2015. For each case TTE, there were two matched spontaneously breathing controls (Control 1: Inpatients scanned in the lab and Control 2: Portable studies). We report the overall mean maximum and view specific quality scores for each TTE. The comprehensiveness of an echo report was calculated as the documented proportion of 12 standard parameters. An inverse probability weighted regression model was fit to determine the relationship between machine learning quality score and the completeness of a TTE report. 175 mechanically ventilated TTEs were included with 350 non-intubated samples (175 Control 1: Lab and 175 Control 2: Portable). In total, the machine learning model analyzed 14,086 echo video clips for quality. The overall accuracy of the model with regard to the expert ground truth for the view classification was 87.0%. The overall mean maximum quality score was lower for mechanically ventilated TTEs (0.55 [95% CI 0.54, 0.56]) versus 0.61 (95% CI 0.59, 0.62) for Control 1: Lab and 0.64 (95% CI 0.63, 0.66) for Control 2: Portable; p = 0.002. Furthermore, mechanically ventilated TTE reports were the least comprehensive, with fewer reported parameters. The regression model demonstrated the correlation of echo image quality and completeness of TTE reporting regardless of the clinical group. Mechanically ventilated TTEs were of inferior quality and clinical utility compared to spontaneously breathing controls and machine learning derived image quality correlates with completeness of TTE reporting regardless of the clinical group.

Artificial Intelligence Aids Cardiac Image Quality Assessment for Improving Precision in Strain Measurements

ObjectivesThe aim of this study was to develop an artificial intelligence tool to assess echocardiographic image quality objectively. BackgroundLeft ventricular global longitudinal strain (LVGLS) has recently been used to monitor cancer therapeutics−related cardiac dysfunction (CTRCD) but image quality limits its reliability. MethodsA DenseNet-121 convolutional neural network was developed for view identification from an athlete’s echocardiographic dataset. To prove the concept that classification confidence (CC) can serve as a quality marker, values of longitudinal strain derived from feature tracking of cardiac magnetic resonance (CMR) imaging and strain analysis of echocardiography were compared. The CC was then applied to patients with breast cancer free from CTRCD to investigate the effects of image quality on the reliability of strain analysis. ResultsCC of the apical 4-chamber view (A4C) was significantly correlated with the endocardial border delineation index. CC of A4C >900 significantly predicted a <15% relative difference in longitudinal strain between CMR feature tracking and automated echocardiographic analysis. Echocardiographic studies (n =752) of 102 patients with breast cancer without CTRCD were investigated. The strain analysis showed higher parallel forms, inter-rater, and test-retest reliabilities in patients with CC of A4C >900. During sequential comparisons of automated LVGLS in individual patients, those with CC of A4C >900 had a lower false positive detection rate of CTRCD. ConclusionsCC of A4C was associated with the reliability of automated LVGLS and could also potentially be used as a filter to select comparable images from sequential echocardiographic studies in individual patients and reduce the false positive detection rate of CTRCD.