Timing and Patient Position During Cuff Blood Pressure Measurement Affect Myocardial Work Parameters Measured by Echocardiography.

BACKGROUND: Although cuff blood pressure measurement is a critical parameter to calculate myocardial work noninvasively, there is no recommendation about when and how to measure it. Accordingly, we sought to evaluate the effects of the timing during the echo study and the patient's position on the scanning bed during the cuff blood pressure measurement on myocardial work parameter calculations. METHODS: One hundred one consecutive patients (44 women, 66 ± 14 years) undergoing clinically indicated echocardiography were prospectively enrolled. During the echocardiographic study, we measured the cuff blood pressure 4 times, using a fully automatic digital blood pressure monitor applied to the right and left arm in the same position throughout the study: BP1, before the start of the echo study, with the patient lying in the supine position; BP2, after positioning the patients on their left side to start the echo study; BP3, at the time of the acquisition of the 3 apical views (4- and 2-chamber and long-axis) used to measured left ventricular global longitudinal strain; and BP4, at the end of the echo study with the patient again in the supine position. RESULTS: Systolic blood pressureat BP1 was 147 ± 21 mm Hg. Between BP1 and BP2, it dropped by 17 ± 9 mm Hg (P < .05). Systolic blood pressure at BP3 was significantly lower than BP2 (130 ± 20 mm Hg vs 122 ± 18 mm Hg, P < .05), and at BP4 was significantly lower than at BP1 (-9 ± 13 mm Hg, P < .05). The average global longitudinal strain was -16% ± 3%. Accordingly, the global work index was 1,929 ± 441 mm Hg% at BP1, dropped to 1,717 ± 421 at BP2, decreased to 1,602 ± 351 mm Hg% at BP3, and increased to 1,815 ± 386 mm Hg% at BP4 (P < .001). CONCLUSIONS: The timing during the echocardiography study and the patient's position on the scanning bed are critical determinants of the measured cuff systolic blood pressure and the resulting values of myocardial work parameters.

DEep LearnIng-based QuaNtification of epicardial adipose tissue predicts MACE in patients undergoing stress CMR.

BACKGROUND AND AIMS: This study investigated the additional prognostic value of epicardial adipose tissue (EAT) volume for major adverse cardiovascular events (MACE) in patients undergoing stress cardiac magnetic resonance (CMR) imaging. METHODS: 730 consecutive patients [mean age: 63 ± 10 years; 616 men] who underwent stress CMR for known or suspected coronary artery disease were randomly divided into derivation (n = 365) and validation (n = 365) cohorts. MACE was defined as non-fatal myocardial infarction and cardiac deaths. A deep learning algorithm was developed and trained to quantify EAT volume from CMR. EAT volume was adjusted for height (EAT volume index). A composite CMR-based risk score by Cox analysis of the risk of MACE was created. RESULTS: In the derivation cohort, 32 patients (8.7 %) developed MACE during a follow-up of 2103 days. Left ventricular ejection fraction (LVEF) < 35 % (HR 4.407 [95 % CI 1.903-10.202]; p<0.001), stress perfusion defect (HR 3.550 [95 % CI 1.765-7.138]; p<0.001), late gadolinium enhancement (LGE) (HR 4.428 [95%CI 1.822-10.759]; p = 0.001) and EAT volume index (HR 1.082 [95 % CI 1.045-1.120]; p<0.001) were independent predictors of MACE. In a multivariate Cox regression analysis, adding EAT volume index to a composite risk score including LVEF, stress perfusion defect and LGE provided additional value in MACE prediction, with a net reclassification improvement of 0.683 (95%CI, 0.336-1.03; p<0.001). The combined evaluation of risk score and EAT volume index showed a higher Harrel C statistic as compared to risk score (0.85 vs. 0.76; p<0.001) and EAT volume index alone (0.85 vs.0.74; p<0.001). These findings were confirmed in the validation cohort. CONCLUSIONS: In patients with clinically indicated stress CMR, fully automated EAT volume measured by deep learning can provide additional prognostic information on top of standard clinical and imaging parameters.

Peak Atrial Longitudinal Strain and Risk Stratification in Moderate and Severe Aortic Stenosis.

AIMS: We sought to investigate the association with outcome of left atrial strain in a large cohort of patients with at least moderate aortic stenosis (AS). METHODS AND RESULTS: we analyzed 467 patients (mean age 80.6 ± 8.2 years; 51% men) with at least moderate AS, and sinus rhythm. The primary study endpoint was the composite of all-cause mortality and hospitalizations for heart failure. After a median follow-up of 19.2 (IQR 12.5-24.4) months, 96 events occurred. Using the ROC curve analysis, the cut-off value of peak atrial longitudinal strain (PALS) more strongly associated with outcome was < 16% [AUC 0.70 (95% CI: 0.63-0.78), p<0.001]. The Kaplan Meier curves demonstrated a higher rate of events for patients with PALS<16% (log-rank p<0.001). On multivariable analysis, PALS [aHR 0.95 (95% CI 0.91 - 0.99), p=0.017] and age were the only variables independently associated with the combined endpoint. PALS provided incremental prognostic value over left ventricular (LV) global longitudinal strain, LV ejection fraction, and right ventricular function. Subgroup analysis revealed that impaired PALS was independently associated with outcome also in the subgroups of paucisymptomatic patients [aHR 0.98 (95% CI 0.97 - 0.98), p=0.048], moderate AS [aHR 0.92, (95% CI 0.86 - 0.98), p=0.016], and low-flow AS [aHR 0.90, (95% CI 0.83 - 0.98), p=0.020]. CONCLUSION: In our patients with at least moderate AS, PALS was independently associated with outcome. In asymptomatic patients, PALS could be a potential marker of subclinical damage, leading to better risk stratification, and, potentially, to earlier treatment.

Clinical Impact of the Volumetric Quantification of Ventricular Secondary Mitral Regurgitation by Three-Dimensional Echocardiography.

BACKGROUND: The assessment of ventricular secondary mitral regurgitation (v-SMR) severity through effective regurgitant orifice area (EROA) and regurgitant volume (RegVol) calculations using the proximal isovelocity surface area (PISA) method and the two-dimensional echocardiography volumetric method (2DEVM) is prone to underestimation. Accordingly, we sought to investigate the accuracy of the three-dimensional echocardiography volumetric method (3DEVM) and its association with outcomes in v-SMR patients. METHODS: We included 229 patients (70 ± 13 years, 74% men) with v-SMR. We compared EROA and RegVol calculated by the 3DEVM, 2DEVM, and PISA methods. The end point was a composite of heart failure hospitalization and death for any cause. RESULTS: After a mean follow-up of 20 ±11 months, 98 patients (43%) reached the end point. Regurgitant volume and EROA calculated by 3DEVM were larger than those calculated by 2DEVM and PISA. Using receiver operating characteristic curve analysis, both EROA (area under the curve, 0.75; 95% CI, 0.68-0.81; P = .008) and RegVol (AUC, 0.75; 95% CI, 0.68-0.82; P = .02) measured by 3DEVM showed the highest association with the outcome at 2 years compared to PISA and 2DEVM (P < .05 for all). Kaplan-Meier analysis demonstrated a significantly higher rate of events in patients with EROA ≥ 0.3 cm2 (cumulative survival at 2 years: 28% ± 7% vs 32% ± 10% vs 30% ± 11%) and RegVol ≥ 45 mL (cumulative survival at 2 years: 21% ± 7% vs 24% ± 13% vs 22% ± 10%) by 3DEVM compared to those by PISA and 2DEVM, respectively. In Cox multivariable analysis, 3DEVM EROA remained independently associated with the end point (hazard ratio, 1.02, 95% CI, 1.00-1.05; P = .02). The model including EROA by 3DEVM provided significant incremental value to predict the combined end point compared to those using 2DEVM (net reclassification index = 0.51, P = .003; integrated discrimination index = 0.04, P = .014) and PISA (net reclassification index = 0.80, P < .001; integrated discrimination index = 0.06, P < .001). CONCLUSIONS: Effective regurgitant orifice area and RegVol calculated by 3DEVM were independently associated with the end point, improving the risk stratification of patients with v-SMR compared to the 2DEVM and PISA methods.

Right Atrial Remodeling and Outcome in Patients with Secondary Tricuspid Regurgitation.

BACKGROUND: In patients with secondary tricuspid regurgitation (STR), right atrial remodeling (RAR) is a proven marker of disease progression. However, the prognostic value of RAR, assessed by indexed right atrial volume (RAVi) and reservoir strain (RAS), remains to be clarified. Accordingly, the aim of our study is to investigate the association with outcome of RAR in patients with STR. METHODS: We enrolled 397 patients (44% men, 72.7 ± 13 years old) with mild to severe STR. Complete two-dimensional and speckle-tracking echocardiography analysis of right atrial and right ventricular (RV) size and function were obtained in all patients. The primary end point was the composite of death from any cause and heart failure hospitalization. RESULTS: After a median follow-up of 15 months (interquartile range, 6-23), the end point was reached by 158 patients (39%). Patients with RAS <13% and RAVi >48 mL/m2 had significantly lower survival rates compared to patients with RAS ≥13% and RAVi ≤48 mL/m2 (log-rank P < .001). On multivariable analysis, RAS <13% (hazard ratio, 2.11; 95% CI, 1.43-3.11; P < .001) and RAVi > 48 mL/m2 (hazard ratio, 1.49; 95% CI, 1.01-2.18; P = .04) remained associated with the combined end point, even after adjusting for RV free-wall longitudinal strain, significant chronic kidney disease, and New York Heart Association class. Secondary tricuspid regurgitation excess mortality increased exponentially with values of 18.2% and 51.3 mL/m2 for RAS and RAVi, respectively. In nested models, the addition of RAS and RAVi provided incremental prognostic value over clinical, conventional echocardiographic parameters of RV size and function and RV free-wall longitudinal strain. CONCLUSIONS: In patients with STR, RAR was independently associated with mortality and heart failure hospitalization. Assessment of RAR could improve risk stratification of patients with STR, potentially identifying those who may benefit from optimization of medical therapy and a closer follow-up.

Circulating Small Extracellular Vesicles Reflect the Severity of Myocardial Damage in STEMI Patients.

Circulating small extracellular vesicles (sEVs) contribute to inflammation, coagulation and vascular injury, and have great potential as diagnostic markers of disease. The ability of sEVs to reflect myocardial damage assessed by Cardiac Magnetic Resonance (CMR) in ST-segment elevation myocardial infarction (STEMI) is unknown. To fill this gap, plasma sEVs were isolated from 42 STEMI patients treated by primary percutaneous coronary intervention (pPCI) and evaluated by CMR between days 3 and 6. Nanoparticle tracking analysis showed that sEVs were greater in patients with anterior STEMI (p = 0.0001), with the culprit lesion located in LAD (p = 0.045), and in those who underwent late revascularization (p = 0.038). A smaller sEV size was observed in patients with a low myocardial salvage index (MSI, p = 0.014). Patients with microvascular obstruction (MVO) had smaller sEVs (p < 0.002) and lower expression of the platelet marker CD41-CD61 (p = 0.039). sEV size and CD41-CD61 expression were independent predictors of MVO/MSI (OR [95% CI]: 0.93 [0.87-0.98] and 0.04 [0-0.61], respectively). In conclusion, we provide evidence that the CD41-CD61 expression in sEVs reflects the CMR-assessed ischemic damage after STEMI. This finding paves the way for the development of a new strategy for the timely identification of high-risk patients and their treatment optimization.

A deep-learning approach for myocardial fibrosis detection in early contrast-enhanced cardiac CT images.

Aims: Diagnosis of myocardial fibrosis is commonly performed with late gadolinium contrast-enhanced (CE) cardiac magnetic resonance (CMR), which might be contraindicated or unavailable. Coronary computed tomography (CCT) is emerging as an alternative to CMR. We sought to evaluate whether a deep learning (DL) model could allow identification of myocardial fibrosis from routine early CE-CCT images. Methods and results: Fifty consecutive patients with known left ventricular (LV) dysfunction (LVD) underwent both CE-CMR and (early and late) CE-CCT. According to the CE-CMR patterns, patients were classified as ischemic (n = 15, 30%) or non-ischemic (n = 35, 70%) LVD. Delayed enhancement regions were manually traced on late CE-CCT using CE-CMR as reference. On early CE-CCT images, the myocardial sectors were extracted according to AHA 16-segment model and labeled as with scar or not, based on the late CE-CCT manual tracing. A DL model was developed to classify each segment. A total of 44,187 LV segments were analyzed, resulting in accuracy of 71% and area under the ROC curve of 76% (95% CI: 72%-81%), while, with the bull's eye segmental comparison of CE-CMR and respective early CE-CCT findings, an 89% agreement was achieved. Conclusions: DL on early CE-CCT acquisition may allow detection of LV sectors affected with myocardial fibrosis, thus without additional contrast-agent administration or radiational dose. Such tool might reduce the user interaction and visual inspection with benefit in both efforts and time.

Cardiac computed tomography: from anatomy to function.

Ischaemic heart disease (IHD) is one of the world's leading causes of morbidity and mortality. Likewise, the diagnosis and risk stratification of patients with coronary artery disease (CAD) have always been based on the detection of the presence and extent of ischaemia by physical or pharmacological stress tests with or without the aid of imaging methods (e.g. exercise stress, test, stress echocardiography, single-photon emission computed tomography, or stress cardiac magnetic resonance). These methods show high performance to assess obstructive CAD, whilst they do not show accurate power to detect non-obstructive CAD. The introduction into clinical practice of coronary computed tomography angiography, the only non-invasive method capable of analyzing the coronary anatomy, allowed to add a crucial piece in the puzzle of the assessment of patients with suspected or chronic IHD. The current review evaluates the technical aspects and clinical experience of coronary computed tomography in the evaluation of atherosclerotic burden with a special focus about the new emerging application such as functional relevance of CAD with fractional flow reserve computed tomography (CT)-derived (FFRct), stress CT perfusion, and imaging inflammatory makers discussing the strength and weakness of each approach.

Quantification of extracellular volume with cardiac computed tomography in patients with dilated cardiomyopathy.

BACKGROUND: Cardiac computed tomography (CCT) was recently validated to measure extracellular volume (ECV) in the setting of cardiac amyloidosis, showing good agreement with cardiovascular magnetic resonance (CMR). However, no evidence is available with a whole-heart single source, single energy CT scanner in the clinical context of newly diagnosed left ventricular dysfunction. Therefore, the aim of this study was to test the diagnostic accuracy of ECVCCT in patients with a recent diagnosis of dilated cardiomyopathy, having ECVCMR as the reference technique. METHODS: 39 consecutive patients with newly diagnosed dilated cardiomyopathy (LVEF <50%) scheduled for clinically indicated CMR were prospectively enrolled. Myocardial segment evaluability assessment with each technique, agreement between ECVCMR and ECVCCT, regression analysis, Bland-Altman analysis and interclass correlation coefficient (ICC) were performed. RESULTS: Mean age of enrolled patients was 62 â€‹± â€‹11 years, and mean LVEF at CMR was 35.4 â€‹± â€‹10.7%. Overall radiation exposure for ECV estimation was 2.1 â€‹± â€‹1.1 â€‹mSv. Out of 624 myocardial segments available for analysis, 624 (100%) segments were assessable by CCT while 608 (97.4%) were evaluable at CMR. ECVCCT demonstrated slightly lower values compared to ECVCMR (all segments, 31.8 â€‹± â€‹6.5% vs 33.9 â€‹± â€‹8.0%, p â€‹< â€‹0.001). At regression analysis, strong correlations were described (all segments, r â€‹= â€‹0.819, 95% CI: 0.791 to 0.844). On Bland-Altman analysis, bias between ECVCMR and ECVCCT for global analysis was 2.1 (95% CI: -6.8 to 11.1). ICC analysis showed both high intra-observer and inter-observer agreement for ECVCCT calculation (0.986, 95%CI: 0.983 to 0.988 and 0.966, 95%CI: 0.960 to 0.971, respectively). CONCLUSIONS: ECV estimation with a whole-heart single source, single energy CT scanner is feasible and accurate. Integration of ECV measurement in a comprehensive CCT evaluation of patients with newly diagnosed dilated cardiomyopathy can be performed with a small increase in overall radiation exposure.

A token-mixer architecture for CAD-RADS classification of coronary stenosis on multiplanar reconstruction CT images.

BACKGROUND AND OBJECTIVE: In patients with suspected Coronary Artery Disease (CAD), the severity of stenosis needs to be assessed for precise clinical management. An automatic deep learning-based algorithm to classify coronary stenosis lesions according to the Coronary Artery Disease Reporting and Data System (CAD-RADS) in multiplanar reconstruction images acquired with Coronary Computed Tomography Angiography (CCTA) is proposed. METHODS: In this retrospective study, 288 patients with suspected CAD who underwent CCTA scans were included. To model long-range semantic information, which is needed to identify and classify stenosis with challenging appearance, we adopted a token-mixer architecture (ConvMixer), which can learn structural relationship over the whole coronary artery. ConvMixer consists of a patch embedding layer followed by repeated convolutional blocks to enable the algorithm to learn long-range dependences between pixels. To visually assess ConvMixer performance, Gradient-Weighted Class Activation Mapping (Grad-CAM) analysis was used. RESULTS: Experimental results using 5-fold cross-validation showed that our ConvMixer can classify significant coronary artery stenosis (i.e., stenosis with luminal narrowing ≥50%) with accuracy and sensitivity of 87% and 90%, respectively. For CAD-RADS 0 vs. 1-2 vs. 3-4 vs. 5 classification, ConvMixer achieved accuracy and sensitivity of 72% and 75%, respectively. Additional experiments showed that ConvMixer achieved a better trade-off between performance and complexity compared to pyramid-shaped convolutional neural networks. CONCLUSIONS: Our algorithm might provide clinicians with decision support, potentially reducing the interobserver variability for coronary artery stenosis evaluation.

Cardiovascular magnetic resonance images with susceptibility artifacts: artificial intelligence with spatial-attention for ventricular volumes and mass assessment.

BACKGROUND: Segmentation of cardiovascular magnetic resonance (CMR) images is an essential step for evaluating dimensional and functional ventricular parameters as ejection fraction (EF) but may be limited by artifacts, which represent the major challenge to automatically derive clinical information. The aim of this study is to investigate the accuracy of a deep learning (DL) approach for automatic segmentation of cardiac structures from CMR images characterized by magnetic susceptibility artifact in patient with cardiac implanted electronic devices (CIED). METHODS: In this retrospective study, 230 patients (100 with CIED) who underwent clinically indicated CMR were used to developed and test a DL model. A novel convolutional neural network was proposed to extract the left ventricle (LV) and right (RV) ventricle endocardium and LV epicardium. In order to perform a successful segmentation, it is important the network learns to identify salient image regions even during local magnetic field inhomogeneities. The proposed network takes advantage from a spatial attention module to selectively process the most relevant information and focus on the structures of interest. To improve segmentation, especially for images with artifacts, multiple loss functions were minimized in unison. Segmentation results were assessed against manual tracings and commercial CMR analysis software cvi42(Circle Cardiovascular Imaging, Calgary, Alberta, Canada). An external dataset of 56 patients with CIED was used to assess model generalizability. RESULTS: In the internal datasets, on image with artifacts, the median Dice coefficients for end-diastolic LV cavity, LV myocardium and RV cavity, were 0.93, 0.77 and 0.87 and 0.91, 0.82, and 0.83 in end-systole, respectively. The proposed method reached higher segmentation accuracy than commercial software, with performance comparable to expert inter-observer variability (bias ± 95%LoA): LVEF 1 ± 8% vs 3 ± 9%, RVEF - 2 ± 15% vs 3 ± 21%. In the external cohort, EF well correlated with manual tracing (intraclass correlation coefficient: LVEF 0.98, RVEF 0.93). The automatic approach was significant faster than manual segmentation in providing cardiac parameters (approximately 1.5 s vs 450 s). CONCLUSIONS: Experimental results show that the proposed method reached promising performance in cardiac segmentation from CMR images with susceptibility artifacts and alleviates time consuming expert physician contour segmentation.

Outcomes and mechanical complications of acute myocardial infarction during the second wave pandemic in a Milan HUB center for cardiac emergencies.

Aims: COVID-19 has dramatically impacted the healthcare system. Evidence from previous studies suggests a decline in in-hospital admissions for acute myocardial infarction (AMI) during the pandemic. However, the effect of the pandemic on mechanical complications (MC) in acute ST-segment elevation myocardial infarction (STEMI) has not been comprehensively investigated. Therefore, we evaluated the impact of the pandemic on MC and in-hospital outcomes in STEMI during the second wave, in which there was a huge SARS-CoV-2 diffusion in Italy. Methods and results: Based on a single center cohort of AMI patients admitted with STEMI between February 1, 2019, and February 28, 2021, we compared the characteristics and outcomes of STEMI patients treated during the pandemic vs. those treated before the pandemic. In total, 479 STEMI patients were included, of which 64.5% were during the pandemic. Relative to before the pandemic, primary percutaneous coronary intervention (PCI) declined (87.7 vs. 94.7%, p = 0.014) during the pandemic. Compared to those admitted before the pandemic (10/2019 to 2/2020), STEMI patients admitted during the second wave (10/2020 to 2/2021) presented with a symptom onset-to-door time greater than 24 h (26.1 vs. 10.3%, p = 0.009) and a reduction of primary PCI (85.2 vs. 97.1%, p = 0.009). MC occurred more often in patients admitted during the second wave of the pandemic than in those admitted before the pandemic (7.0 vs. 0.0%, p = 0.032). In-hospital mortality increased during the second wave (10.6 vs. 2.9%, p = 0.058). Conclusion: Although the experience gained during the first wave and a more advanced hub-and-spoke system for cardiovascular emergencies persists, late hospitalizations and a high incidence of mechanical complications in STEMI were observed even in the second wave.

Comparison between Automatic and Semiautomatic System for the 3D Echocardiographic Multiparametric Evaluation of RV Function and Dimension.

BACKGROUND: The right ventricle (RV) plays a pivotal role in cardiovascular diseases and 3-dimensional echocardiography (3DE) has gained acceptance for the evaluation of RV volumes and function. Recently, a new artificial intelligence (AI)-based automated 3DE software for RV evaluation has been proposed and validated against cardiac magnetic resonance. The aims of this study were three-fold: (i) feasibility of the AI-based 3DE RV quantification, (ii) comparison with the semi-automatic 3DE method and (iii) assessment of 2-dimensional echocardiography (2DE) and strain measurements obtained automatically. METHODS: A total of 203 subject (122 normal and 81 patients) underwent a 2DE and both the semi-automatic and automatic 3DE methods for Doppler standard, RV volumes and ejection fraction (RVEF) measurements. RESULTS: The automatic 3DE method was highly feasible, faster than 2DE and semi-automatic 3DE and data obtained were comparable with traditional measurements. Both in normal subjects and patients, the RVEF was similar to the two 3DE methods and 2DE and strain measurements obtained by the automated system correlated very well with the standard 2DE and strain ones. CONCLUSIONS: results showed that rapid analysis and excellent reproducibility of AI-based 3DE RV analysis supported the routine adoption of this automated method in the daily clinical workflow.

Decision Support Systems in HF based on Deep Learning Technologies.

PURPOSE OF REVIEW: Application of deep learning (DL) is growing in the last years, especially in the healthcare domain. This review presents the current state of DL techniques applied to electronic health record structured data, physiological signals, and imaging modalities for the management of heart failure (HF), focusing in particular on diagnosis, prognosis, and re-hospitalization risk, to explore the level of maturity of DL in this field. RECENT FINDINGS: DL allows a better integration of different data sources to distillate more accurate outcomes in HF patients, thus resulting in better performance when compared to conventional evaluation methods. While applications in image and signal processing for HF diagnosis have reached very high performance, the application of DL to electronic health records and its multisource data for prediction could still be improved, despite the already promising results. Embracing the current big data era, DL can improve performance compared to conventional techniques and machine learning approaches. DL algorithms have potential to provide more efficient care and improve outcomes of HF patients, although further investigations are needed to overcome current limitations, including results generalizability and transparency and explicability of the evidences supporting the process.

Machine Learning Prediction Models for Mitral Valve Repairability and Mitral Regurgitation Recurrence in Patients Undergoing Surgical Mitral Valve Repair.

BACKGROUND: Mitral valve regurgitation (MR) is the most common valvular heart disease and current variables associated with MR recurrence are still controversial. We aim to develop a machine learning-based prognostic model to predict causes of mitral valve (MV) repair failure and MR recurrence. METHODS: 1000 patients who underwent MV repair at our institution between 2008 and 2018 were enrolled. Patients were followed longitudinally for up to three years. Clinical and echocardiographic data were included in the analysis. Endpoints were MV repair surgical failure with consequent MV replacement or moderate/severe MR (>2+) recurrence at one-month and moderate/severe MR recurrence after three years. RESULTS: 817 patients (DS1) had an echocardiographic examination at one-month while 295 (DS2) also had one at three years. Data were randomly divided into training (DS1: n = 654; DS2: n = 206) and validation (DS1: n = 164; DS2 n = 89) cohorts. For intra-operative or early MV repair failure assessment, the best area under the curve (AUC) was 0.75 and the complexity of mitral valve prolapse was the main predictor. In predicting moderate/severe recurrent MR at three years, the best AUC was 0.92 and residual MR at six months was the most important predictor. CONCLUSIONS: Machine learning algorithms may improve prognosis after MV repair procedure, thus improving indications for correct candidate selection for MV surgical repair.

Head to Head Comparison between Different 3-Dimensional Echocardiographic Rendering Tools in the Imaging of Percutaneous Edge-to-Edge Mitral Valve Repair.

MitraClip (MC) is the most common percutaneous treatment for severe mitral regurgitation (MR). An accurate two-dimensional and three-dimensional echocardiographic (3DTEE) imaging is mandatory for the optimal procedural result. Recently transillumination 3DTEE rendering (3DTr) has been introduced integrating a virtual light source into the dataset and with the addition of glass effect (3DGl) allows to adjust tissue transparency improving depth perception and anatomical structure delineation in comparison with the standard 3DTEE (3DSt). The aim of this retrospective study in 30 patients undergoing MC, was to compare 3DSt, 3DTr, and 3DGl in mitral valve (MV) evaluation and procedural result assessment. 3DTEE acquisitions obtained before and after MC were processed with 3DSt, 3DTr, and 3DGl rendering. Each reconstruction was scored for quality and for ability to recognize MV anatomy, MR origin, clip position, dimension and grasping. Imaging quality was judged good or optimal in 52%, 76%, and 96% in 3DSt, 3DTr, and 3DGl reconstructions respectively. In 26/30 patients a diagnostic incremental value was found with 3DTr vs. 3DSt and in 15/26 with 3DGl vs. 3DTr and 3DSt. Only 3DGl with perpendicular cropping of the clip allowed to visualize and measure the grasped portion of each mitral leaflets. 3DTEE imaging during MC may be improved by 3DTr and 3DGl providing a better evaluation of MV, of leaflet grasping and of residual MR jets after MC.

Predicting Long-Term Mortality in TAVI Patients Using Machine Learning Techniques.

BACKGROUND: Whereas transcatheter aortic valve implantation (TAVI) has become the gold standard for aortic valve stenosis treatment in high-risk patients, it has recently been extended to include intermediate risk patients. However, the mortality rate at 5 years is still elevated. The aim of the present study was to develop a novel machine learning (ML) approach able to identify the best predictors of 5-year mortality after TAVI among several clinical and echocardiographic variables, which may improve the long-term prognosis. METHODS: We retrospectively enrolled 471 patients undergoing TAVI. More than 80 pre-TAVI variables were collected and analyzed through different feature selection processes, which allowed for the identification of several variables with the highest predictive value of mortality. Different ML models were compared. RESULTS: Multilayer perceptron resulted in the best performance in predicting mortality at 5 years after TAVI, with an area under the curve, positive predictive value, and sensitivity of 0.79, 0.73, and 0.71, respectively. CONCLUSIONS: We presented an ML approach for the assessment of risk factors for long-term mortality after TAVI to improve clinical prognosis. Fourteen potential predictors were identified with the organic mitral regurgitation (myxomatous or calcific degeneration of the leaflets and/or annulus) which showed the highest impact on 5 years mortality.

Automated left and right ventricular chamber segmentation in cardiac magnetic resonance images using dense fully convolutional neural network.

BACKGROUND AND OBJECTIVE: Segmentation of the left ventricular (LV) myocardium (Myo) and RV endocardium on cine cardiac magnetic resonance (CMR) images represents an essential step for cardiac-function evaluation and diagnosis. In order to have a common reference for comparing segmentation algorithms, several CMR image datasets were made available, but in general they do not include the most apical and basal slices, and/or gold standard tracing is limited to only one of the two ventricles, thus not fully corresponding to real clinical practice. Our aim was to develop a deep learning (DL) approach for automated segmentation of both RV and LV chambers from short-axis (SAX) CMR images, reporting separately the performance for basal slices, together with the applied criterion of choice. METHOD: A retrospectively selected database (DB1) of 210 cine sequences (3 pathology groups) was considered: images (GE, 1.5 T) were acquired at Centro Cardiologico Monzino (Milan, Italy), and end-diastolic (ED) and end-systolic frames (ES) were manually segmented (gold standard, GS). Automatic ED and ES RV and LV segmentation were performed with a U-Net inspired architecture, where skip connections were redesigned introducing dense blocks to alleviate the semantic gap between the U-Net encoder and decoder. The proposed architecture was trained including: A) the basal slices where the Myo surrounded the LV for at least the 50% and all the other slice; B) all the slices where the Myo completely surrounded the LV. To evaluate the clinical relevance of the proposed architecture in a practical use case scenario, a graphical user interface was developed to allow clinicians to revise, and correct when needed, the automatic segmentation. Additionally, to assess generalizability, analysis of CMR images obtained in 12 healthy volunteers (DB2) with different equipment (Siemens, 3T) and settings was performed. RESULTS: The proposed architecture outperformed the original U-Net. Comparing the performance on DB1 between the two criteria, no significant differences were measured when considering all slices together, but were present when only basal slices were examined. Automatic and manually-adjusted segmentation performed similarly compared to the GS (bias±95%LoA): LVEDV -1±12 ml, LVESV -1±14 ml, RVEDV 6±12 ml, RVESV 6±14 ml, ED LV mass 6±26 g, ES LV mass 5±26 g). Also, generalizability showed very similar performance, with Dice scores of 0.944 (LV), 0.908 (RV) and 0.852 (Myo) on DB1, and 0.940 (LV), 0.880 (RV), and 0.856 (Myo) on DB2. CONCLUSIONS: Our results support the potential of DL methods for accurate LV and RV contours segmentation and the advantages of dense skip connections in alleviating the semantic gap generated when high level features are concatenated with lower level feature. The evaluation on our dataset, considering separately the performance on basal and apical slices, reveals the potential of DL approaches for fast, accurate and reliable automated cardiac segmentation in a real clinical setting.

Novelties in 3D Transthoracic Echocardiography.

Cardiovascular imaging is developing at a rapid pace and the newer modalities, in particular three-dimensional echocardiography, allow better analysis of heart structures. Identifying valve lesions and grading their severity represents crucial information and nowadays is strengthened by the introduction of new software, such as transillumination, which provide detailed morphology descriptions. Chambers quantification has never been so rapid and accurate: machine learning algorithms generate automated volume measurements, including left ventricular systolic and diastolic function, which is extremely important for clinical decisions. This review provides an overview of the latest innovations in the echocardiography field, and is helpful by providing a better insight into heart diseases.