Correcting bias in cardiac geometries derived from multimodal images using spatiotemporal mapping.

Cardiovascular imaging studies provide a multitude of structural and functional data to better understand disease mechanisms. While pooling data across studies enables more powerful and broader applications, performing quantitative comparisons across datasets with varying acquisition or analysis methods is problematic due to inherent measurement biases specific to each protocol. We show how dynamic time warping and partial least squares regression can be applied to effectively map between left ventricular geometries derived from different imaging modalities and analysis protocols to account for such differences. To demonstrate this method, paired real-time 3D echocardiography (3DE) and cardiac magnetic resonance (CMR) sequences from 138 subjects were used to construct a mapping function between the two modalities to correct for biases in left ventricular clinical cardiac indices, as well as regional shape. Leave-one-out cross-validation revealed a significant reduction in mean bias, narrower limits of agreement, and higher intraclass correlation coefficients for all functional indices between CMR and 3DE geometries after spatiotemporal mapping. Meanwhile, average root mean squared errors between surface coordinates of 3DE and CMR geometries across the cardiac cycle decreased from 7 ± 1 to 4 ± 1 mm for the total study population. Our generalised method for mapping between time-varying cardiac geometries obtained using different acquisition and analysis protocols enables the pooling of data between modalities and the potential for smaller studies to leverage large population databases for quantitative comparisons.

MITEA: A dataset for machine learning segmentation of the left ventricle in 3D echocardiography using subject-specific labels from cardiac magnetic resonance imaging.

Segmentation of the left ventricle (LV) in echocardiography is an important task for the quantification of volume and mass in heart disease. Continuing advances in echocardiography have extended imaging capabilities into the 3D domain, subsequently overcoming the geometric assumptions associated with conventional 2D acquisitions. Nevertheless, the analysis of 3D echocardiography (3DE) poses several challenges associated with limited spatial resolution, poor contrast-to-noise ratio, complex noise characteristics, and image anisotropy. To develop automated methods for 3DE analysis, a sufficiently large, labeled dataset is typically required. However, ground truth segmentations have historically been difficult to obtain due to the high inter-observer variability associated with manual analysis. We address this lack of expert consensus by registering labels derived from higher-resolution subject-specific cardiac magnetic resonance (CMR) images, producing 536 annotated 3DE images from 143 human subjects (10 of which were excluded). This heterogeneous population consists of healthy controls and patients with cardiac disease, across a range of demographics. To demonstrate the utility of such a dataset, a state-of-the-art, self-configuring deep learning network for semantic segmentation was employed for automated 3DE analysis. Using the proposed dataset for training, the network produced measurement biases of -9 ± 16 ml, -1 ± 10 ml, -2 ± 5 %, and 5 ± 23 g, for end-diastolic volume, end-systolic volume, ejection fraction, and mass, respectively, outperforming an expert human observer in terms of accuracy as well as scan-rescan reproducibility. As part of the Cardiac Atlas Project, we present here a large, publicly available 3DE dataset with ground truth labels that leverage the higher resolution and contrast of CMR, to provide a new benchmark for automated 3DE analysis. Such an approach not only reduces the effect of observer-specific bias present in manual 3DE annotations, but also enables the development of analysis techniques which exhibit better agreement with CMR compared to conventional methods. This represents an important step for enabling more efficient and accurate diagnostic and prognostic information to be obtained from echocardiography.

Inequality and inequity in access to health care and treatment for chronic conditions in China: the Guangzhou Biobank Cohort Study.

Non-communicable diseases (NCDs) are a large and rapidly-growing problem in China and other middle-income countries. Clinical treatment of NCDs is long-term and expensive, so it may present particular problems for equality and horizontal equity (equal treatment for equal need) in access to health care, although little is known about this at present in low- and middle-income countries. To address this gap, and inform policy for a substantial proportion of the global population, we examined inequality and inequity in general health care utilization (doctor consultations and hospital admissions) and in treatment of chronic conditions (hypertension, hyperglycaemia and dyslipidaemia), in 30 499 Chinese adults aged ≥50 years from one of China's richest provinces, using the Guangzhou Biobank Cohort Study (2003-2008). We used concentration indices to test for inequality and inequity in utilization by household income per head. Inequality was decomposed to show the contributions of income, indicators of 'need for health care' (age, sex, self-rated health, coronary heart disease risk and chronic obstructive pulmonary disease) and non-need factors (education, occupation, out-of-pocket health care payments and health insurance). We found inequality and inequity in treatment of chronic conditions but not in general health care utilization. Using more objective and specific measures of 'need for health care' increased estimates of inequity for treatment of chronic conditions. Income and non-need factors (especially health insurance, education and occupation) made the largest contributions to inequality. Further work is needed on why access to treatment for chronic conditions in China is restricted for those on low incomes and how these inequities can be mitigated.

Life-course origins of social inequalities in adult immune cell markers of inflammation in a developing southern Chinese population: the Guangzhou Biobank Cohort Study.

BACKGROUND: Socioeconomic position (SEP) throughout life is associated with cardiovascular disease, though the mechanisms linking these two are unclear. It is also unclear whether there are critical periods in the life course when exposure to better socioeconomic conditions confers advantages or whether SEP exposures accumulate across the whole life course. Inflammation may be a mechanism linking socioeconomic position (SEP) with cardiovascular disease. In a large sample of older residents of Guangzhou, in southern China, we examined the association of life course SEP with inflammation. METHODS: In baseline data on 9,981 adults (≥ 50 years old) from the Guangzhou Biobank Cohort Study (2006-08), we used multivariable linear regression and model fit to assess the associations of life course SEP at four stages (childhood, early adult, late adult and current) with white blood, granulocyte and lymphocyte cell counts. RESULTS: A model including SEP at all four life stages best explained the association of life course SEP with white blood and granulocyte cell count for men and women, with early adult SEP (education) making the largest contribution. A critical period model best explained the association of life course SEP with lymphocyte count, with sex-specific associations. Early adult SEP was negatively associated with lymphocytes for women. CONCLUSIONS: Low SEP throughout life may negatively impact late adult immune-inflammatory status. However, some aspects of immune-inflammatory status may be sensitive to earlier exposures, with sex-specific associations. The findings were compatible with the hypothesis that in a developing population, upregulation of the gonadotropic axis with economic development may obscure the normally protective effects of social advantage for men.

Socioeconomic influences at different life stages on health in Guangzhou, China.

In long-term developed countries socioeconomic position across the life course is positively associated with health. We examined these associations in a developing country with a history of efforts to reorganize the social hierarchy. Taking a life course perspective, we used multi-variable logistic regression to assess the association of socioeconomic disadvantage at four life stages (measured by parental possessions, education, longest-held occupation and current household income) with self-rated health, chronic obstructive pulmonary disease (COPD) and metabolic syndrome in 20,086 Chinese adults aged ≥50 years from the Guangzhou Biobank Cohort Study (2005-2008). Model comparisons were used to determine whether the number of exposures to disadvantage (accumulation of risk) was more important than the life stage of exposure (critical periods). Socioeconomic disadvantage across the life course was associated with poor self-rated health, COPD and, in women only, with metabolic syndrome. Adjusting for adult health-related behavior (smoking, alcohol use and physical exercise) altered these associations very little. Associations between socioeconomic disadvantage and health in this Southern Chinese population were broadly similar to those found in Western countries in terms of the accumulation of disadvantage across the life course. However, longest-held occupation was not independently associated with adult health and socioeconomic disadvantage was not associated with metabolic syndrome in men. This suggests that the mechanisms linking socioeconomic position to health in China may be different from those in Western populations and may require context-specific policy interventions.

Plasma natriuretic peptide levels increase with symptoms and severity of mitral regurgitation.

OBJECTIVES: This paper will describe associations between plasma natriuretic peptide levels and the severity and symptoms of mitral regurgitation (MR). BACKGROUND: A biochemical test that assisted grading of the severity of MR and the interpretation of symptoms would be of clinical value. METHODS: Forty-nine patients with isolated MR and left ventricular (LV) ejection fractions (EFs) of >55% underwent transthoracic echocardiography, assessment of symptoms, and measurement of atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and its amino-terminal portion, N-BNP. RESULTS: The level of each natriuretic peptide rose with increasing severity of MR and with increases in left atrial (LA) dimensions (p < 0.001 for all comparisons), but no significant correlation existed between any natriuretic peptide and the LV dimensions or EF. Natriuretic peptide levels were higher in symptomatic MR (n = 16, BNP geometric mean 16.9 [95% confidence interval (CI) 13.3 to 21.4] pmol/l) compared with asymptomatic MR (n = 33, BNP 7.1 [95% CI 6.0 to 8.4] pmol/l, p < 0.001), and higher in asymptomatic MR than in normal controls (n = 100, BNP 5.3 [95% CI 4.8 to 5.8] pmol/l, p < 0.0001). These differences were similar for N-BNP and ANP and remained statistically significant (p < 0.05) after adjustment for echocardiographic measures of LV function and severity of MR. Both the sensitivity and the specificity for symptoms for the natriuretic peptides (area under receiver-operator characteristic curve for BNP = 0.90, N-BNP = 0.89, ANP = 0.89) were similar to the MR score (0.88) and greater than for LA dimension (0.81), vena contracta width (0.82), and LV end-systolic dimension (0.63). CONCLUSIONS: Plasma natriuretic peptides levels increase with the severity of MR and are higher in symptomatic compared to asymptomatic patients, even when LV EF is normal.

Increased plasma natriuretic peptide levels reflect symptom onset in aortic stenosis.

BACKGROUND: The onset of symptoms is a critical point in the natural history of aortic stenosis and the cardinal indication for valve replacement. This study assessed the associations between natriuretic peptide levels, disease severity, and cardiac symptoms in aortic stenosis. METHODS AND RESULTS: Seventy-four patients with isolated aortic stenosis underwent independent assessment of symptoms, transthoracic echocardiography, and measurement of plasma levels of atrial natriuretic peptide, brain natriuretic peptide (BNP), and N-BNP. Natriuretic peptide levels were also measured in 100 clinically normal control subjects. The aortic valve area was smaller in symptomatic patients (n=45) than in asymptomatic patients (n=29; mean, 0.71+/-0.23 cm2 and 0.99+/-0.31 cm2, respectively; P<0.0001). Plasma natriuretic peptide levels were higher in symptomatic patients than in asymptomatic patients (for N-BNP: median, 112 versus 33 pmol/L; interquartile range, 70 to 193 versus 16 to 58 pmol/L, respectively; P=0.0002). After adjustment for age, sex, serum creatinine, aortic valve area, and left ventricular ejection fraction, N-BNP levels were 1.74 times higher (95% confidence interval, 1.12 to 2.69) for symptomatic than asymptomatic patients with aortic stenosis (P=0.014). Natriuretic peptide levels increased with the New York Heart Association class (for N-BNP median values were 13, 34, 105, and 202 pmol/L for normal control subjects, class I, class II, and class III/IV patients, respectively; interquartile ranges for the same patients were 8 to 21, 16 to 58, 57 to 159, and 87 to 394 pmol/L; P<0.0001). Similar associations were observed for BNP and atrial natriuretic peptide. CONCLUSIONS: Plasma natriuretic peptide levels are elevated in symptomatic patients with aortic stenosis. Measurement of natriuretic peptides may complement clinical and echocardiographic evaluation of patients with aortic stenosis.

Potential thrombophilic mutations/polymorphisms in patients with no flow-limiting stenosis after myocardial infarction.

BACKGROUND: Although inherited thrombophilias are more common in patients with venous thromboembolism, their influence on the development of myocardial infarction (MI) requires clarification. METHODS AND RESULTS: To determine whether there are increased frequencies of mutations/polymorphisms in 14 genes potentially causing thrombophilia in patients with no flow-limiting stenoses after MI compared with patients with > or =1 flow-limiting stenosis of >50%, we studied 395 patients (60 with no flow-limiting stenosis) who underwent angiography at approximately 1 month. The mutations/polymorphisms studied included Factor V Leiden, prothrombin variant G20210A, beta-fibrinogen 448 (G/A), endothelial protein C receptor (23-base pair insertion), methyl tetrahydrofolate reductase 677 (C/T), platelet glycoprotein IIIa PlA1/A2, plasminogen activator inhibitor-1 4G/5G, angiotensin II type 1 receptor (A/C), hemochromatosis gene 282 (G/A), nitric oxide synthase (NOS) (3 forms: eNOS, eNOS3, eNOS4), p22 phox of NADPH oxidase C242T, and angiotensin-converting enzyme insertion/deletion polymorphism. The frequencies of Factor V Leiden and the beta-fibrinogen 448 A allele were higher in patients with no flow-limiting stenosis than in patients with > or =1 stenosis (11.7% vs 3.6%, odds ratio [OR] 3.6, 95% CI 1.3-9.4, P =.015; and 42% vs 27%, OR 2.0, 95% CI 1.1-3.5, P =.018, respectively), and there was a trend toward an increased frequency of prothrombin variant G20210A (6.7% vs 2.1%, OR 3.4, 95% CI 0.95-11.8, P =.069). However, in patients with no flow-limiting stenosis after MI the frequencies of the other gene mutations/polymorphisms were not increased. Also, there were no significant interactions between any of these 14 mutation/polymorphisms, major cardiovascular risk factors, and the absence of any flow-limiting stenosis, except for Factor V Leiden and hypertension (OR 6.34, 95% CI 2.67-100, P =.004). CONCLUSIONS: Patients with no flow-limiting stenosis after MI had increased frequencies of 2 inherited thrombophilias (Factor V Leiden and beta-fibrinogen 448 A allele), and there was a trend toward an increased frequency of prothrombin variant G20210A compared with patients with > or =1 stenosis. These data suggest that polymorphisms/mutations in some gene products influencing coagulation may influence the pathogenesis of MI.