Unraveling diagnostic co-morbidity makeup of each HF category as characteristically derived by ECG- and ECHO-findings, a prevalence analysis

Heart Failure (HF) relies mainly on measurements from Echocardiography, in particular Echo-Findings, to estimate Left Ventricle Ejection Fraction (LVEF) and evaluating structural heart disease criteria. As Echocardiography is not available in primary care, the key structural (heart chamber enlargements) and functional abnormality related measurements are not available precluding the ability to diagnose HF other than through mainly symptomatic means. The opportunity for earlier detection of HF is lost.In this work, we first explore each of the three HF categories, preserved EF, mild-reduced EF, and reduced EF, using various morphological and functional etiology-specific characteristics supported by a literature review and an extensive analysis of a large, dedicated database accumulated over 8 years.We then explore the typical signs and co-morbidities of HF using prevalence analysis to unravel the diagnostic makeup of each HF category as characteristically derived by ECG- and ECHO-findings. From this, we then conduct a principal component analysis (PCA) of the data to interpret patterns of comorbidities, showing groups of comorbidities frequently associated with each other.Lastly, we delve into the role of breakthrough methods for the analysis of bio-signals to replicate common ECHO-findings, as alternatives for detecting and diagnosing HF similarly to Echocardiography, thereby providing a simple device for the effective detection of HF for use in Primary Care.

Effect of the number of parity on right heart chamber quantification

Introduction Pregnancy is a process that causes several physiological changes including all systems as well as cardiovascular system. Ventricular hypertrophy and dilation of cardiac chambers are seen as a result of these changes. Although there are studies about pregnancy-related changes in echocardiographic examination; there is no data about the long-term effects of parity on these alterations. Therefore, we have evaluated the long-term effect of pregnancy on right ventricular (RV) dilation and RV hypertrophy and their relation to the parity number. Methods This prospective study included a total of 600 women (200 consecutive women who had no parity, 200 women who had a parity number of 1 to 4 and 200 women who had a parity number of more than 4). Right chambers’ measurements were compared between the groups. Results In echocardiographic analysis, RV and right atrial dimensions and areas and RV wall thickness were higher in parous women. On the other hand, RV systolic function parameters were significantly lower in parous women. These significant changes showed a gradual increase or decrease by increasing parity number. There were also independent relationship between the number of parity and RV hypertrophy even after adjustment for several confounders. Conclusion Pregnancy-related physiological changes mostly resolve after delivery. This study about long-term effects of pregnancy on RV has demonstrated that there is a significant relation between the number of parity and either RV dilation or RV hypertrophy. Each parity had also additive effect on these changes.

BOOSTING SEGMENTATION ACCURACY OF THE DEEP LEARNING MODELS BASED ON THE SYNTHETIC DATA GENERATION

In the era of data-driven machine learning algorithms, data represents a new oil. The application of machine learning algorithms shows they need large heterogeneous datasets that crucially are correctly labeled. However, data collection and its labeling are time-consuming and labor-intensive processes. A particular task we solve using machine learning is related to the segmentation of medical devices in echocardiographic images during minimally invasive surgery. However, the lack of data motivated us to develop an algorithm generating synthetic samples based on real datasets. The concept of this algorithm is to place a medical device (catheter) in an empty cavity of an anatomical structure, for example, in a heart chamber, and then transform it. To create random transformations of the catheter, the algorithm uses a coordinate system that uniquely identifies each point regardless of the bend and the shape of the object. It is proposed to take a cylindrical coordinate system as a basis, modifying it by replacing the Z-axis with a spline along which the h-coordinate is measured. Having used the proposed algorithm, we generated new images with the catheter inserted into different heart cavities while varying its location and shape. Afterward, we compared the results of deep neural networks trained on the datasets comprised of real and synthetic data. The network trained on both real and synthetic datasets performed more accurate segmentation than the model trained only on real data. For instance, modified U-net trained on combined datasets performed segmentation with the Dice similarity coefficient of 92.6±2.2%, while the same model trained only on real samples achieved the level of 86.5±3.6%. Using a synthetic dataset allowed decreasing the accuracy spread and improving the generalization of the model. It is worth noting that the proposed algorithm allows reducing subjectivity, minimizing the labeling routine, increasing the number of samples, and improving the heterogeneity.

Genetic Analysis of Right Heart Structure and Function in 40,000 People

The heart evolved hundreds of millions of years ago. During mammalian evolution, the cardiovascular system developed with complete separation between pulmonary and systemic circulations incorporated into a single pump with chambers dedicated to each circulation. A lower pressure right heart chamber supplies deoxygenated blood to the lungs, while a high pressure left heart chamber supplies oxygenated blood to the rest of the body. Due to the complexity of morphogenic cardiac looping and septation required to form these two chambers, congenital heart diseases often involve maldevelopment of the evolutionarily recent right heart chamber. Additionally, some diseases predominantly affect structures of the right heart, including arrhythmogenic right ventricular cardiomyopathy (ARVC) and pulmonary hypertension. To gain insight into right heart structure and function, we fine-tuned deep learning models to recognize the right atrium, the right ventricle, and the pulmonary artery, and then used those models to measure right heart structures in over 40,000 individuals from the UK Biobank with magnetic resonance imaging. We found associations between these measurements and clinical disease including pulmonary hypertension and dilated cardiomyopathy. We then conducted genome-wide association studies, identifying 104 distinct loci associated with at least one right heart measurement. Several of these loci were found near genes previously linked with congenital heart disease, such as NKX2-5, TBX3, WNT9B, and GATA4. We also observed interesting commonalities and differences in association patterns at genetic loci linked with both right and left ventricular measurements. Finally, we found that a polygenic predictor of right ventricular end systolic volume was associated with incident dilated cardiomyopathy (HR 1.28 per standard deviation; P = 2.4E-10), and remained a significant predictor of disease even after accounting for a left ventricular polygenic score. Harnessing deep learning to perform large-scale cardiac phenotyping, our results yield insights into the genetic and clinical determinants of right heart structure and function.

Left atrial strain rate during atrial contraction predicts raised pulmonary capillary wedge pressure: evidence for left atrio-ventricular interaction

This study aimed to assess the relationship between different LA strain components and PCWP as well as to the relationship with other established methods. We studied 144 symptomatic patients, age 63 ± 14 years, 54 males, using conventional transthoracic echocardiography protocols, including LA and LV myocardial deformation from speckle tracking technique investigations along with simultaneous right heart catheterization (RHC) using established techniques. From RHC, pulmonary artery pressure (PAP), and pulmonary capillary wedge pressure (PCWP) were measured and pulmonary vascular resistance (PVR) calculated. LA strain rate during atrial contraction (LASRa) was the strongest correlate with PCWP (r2 = − 0.40, p < 0.001), over and above both LASR during LV systole (LASRs) and LA longitudinal strain during ventricular systole (LASs) (r2 = 0.21 and 0.19, respectively, p < 0.001 for both). The correlation between LASRa and PCWP was stronger in patients with post-capillary PH compared to pre-capillary PH (r2 = 0.21 vs. r2 = 0.02, respectively). The strongest relationship between LASRa and PCWP was in patients with enlarged LA volume > 34 ml/m2 (r2 = 0.60, p < 0.001). In all patients LASRa < = 0.9 1/s was 88% accurate in predicting LA pressure > 15 mmHg which was superior to recently proposed uni- and multi-variable models. LASR during atrial contraction is the strongest predictor of PCWP, particularly in patients with post-capillary PH and with dilated LA cavity. Furthermore, it proved superior to recently proposed uni- and multi-variable based algorithms. Its close relationship with LV strain rate counterpart reflects important left heart chamber interaction in patients with raised LA pressure.

Analysis of vortex ring formation in the heart chamber by instantaneous vortex deviation based on convolutional neural network

The formation of vortex rings during the left ventricle (LV) filling is an optimized mechanism for blood transport, and the vorticity is an important measure of a healthy heart and LV. There is a relationship between abnormal diastolic vortex structure and impaired LV, and hence vortex identification is vital for understanding the underlying physical mechanism of blood flow. However, due to lack of quantitative methods, defining, computing and mapping the left ventricular vortices has not been rigorously studied previously. In this paper, a novel method of vortex detection based on the convolutional neural network (CNN) is created, which enables determination of the boundary of vortex and integrates the local and global flow fields. We have used the CNN-based vortex identification and vector flow mapping (VFM) to quantify left ventricular vorticity. In the clinical application of our methodology to healthy subjects and uremic patients, we find differences in the strength and position of the vortices between healthy and patients with uremia cardiomyopathy. Our results can accurately indicate the role of vortex formation in intracardiac flow, and provide new insights into the blood flow within the heart structure.

Abstract 15561: Incremental Prognostic Value of Biventricular Longitudinal Strain and Hs-tni in Covid-19 Patients

Background: Myocardial strain derived from two-dimensional speckle-tracking echocardiography (2D-STE) has been shown to be more sensitive to detect early ventricular dysfunction than conventional echocardiography. However, the study about the prognostic value of biventricular longitudinal strain in coronavirus disease 2019 (COVID-19) is still scarce. Aims: We aimed to evaluate the prognostic value of biventricular longitudinal strain and its combination with high-sensitivity troponin I (hs-TNI) in COVID-19 patients. Methods: We enrolled a total of 160 COVID-19 patients who underwent both echocardiogram and hs-TNI testing. The cardiac structure, function and myocardial strain were compared between patients with and without elevated hs-TNI levels. Left ventricular longitudinal strain (LV LS) and right ventricular free wall longitudinal strain (RVFWLS) were determined by 2D-STE. Results: Compared with patients with normal hs-TNI levels, patients with elevated hs-TNI levels had diminished LV diastolic function, larger right-heart chamber, higher proportion of pulmonary hypertension, lower LV LS and RVFWLS. During a median follow-up of 60 days, 23 patients died. The multivariant analysis revealed LV LS and RVFWLS [Odd ratio (confidence interval): 1.533 (1.131-2.079); P =0.006; 1.267 (1.101-1.794), P =0.021, respectively] both were the independent predictors of higher mortality. Further, receiver-operating characteristic analysis revealed that the accuracy for predicting death was greater for the combination of hs-TNI levels with LV LS than separate LV LS (AUC: 0.93 vs 0.77, P =0.001), and for the combination of hs-TNI levels with RVFWLS than RVFWLS alone (AUC: 0.92 vs 0.83, P =0.041). Conclusion: Our study highlights that the combination of ventricular longitudinal strain with hs-TNI can provide a higher accuracy for predicting mortality in COVID-19 patients, which may enhance risk stratification in COVID-19 patients.