scholarly journals Associations between plasma metabolite profiles and blood pressure: the HELIUS study

2021 ◽  
Vol 42 (Supplement_1) ◽  
Author(s):  
B Verhaar ◽  
C M Mosterd ◽  
D Collard ◽  
H Galenkamp ◽  
B J H Van Den Born ◽  
...  
Keyword(s):  
Machine Learning ◽  
Blood Pressure ◽  
Regression Models ◽  
African Descent ◽  
Plasma Metabolites ◽  
Linear Regression Models ◽  
Helius Study ◽  
Metabolite Profiles ◽  
Catecholamine Metabolites ◽  
Plasma Metabolite

Abstract Background Blood pressure (BP) is regulated by plasma metabolites from different neurohumoral and cardiometabolic systems. Since there are established differences in hypertension pathogenesis and treatment response between ethnicities, we hypothesized that plasma metabolites may be differently associated with BP across ethnic groups. Purpose To investigate associations between plasma metabolite profiles and BP in a multi-ethnic population-based cohort. Methods From the Healthy Living In an Urban Setting (HELIUS) study, 369 subjects (mean age 52±11 years, 51%F) of African and non-African descent were included. Office systolic (136±21 mmHg) and diastolic (83±12 mmHg) BP levels were recorded. Plasma metabolites were measured semi-quantitively with LC-MS (Metabolon) from fasting plasma samples. Associations between metabolite profiles and BP were assessed with machine learning prediction models using the XGBoost algorithm with nested cross-validation. Associations between the resulting best predictors and BP were assessed with linear regression models while adjusting for age, sex, estimated glomerular filtration rate and diabetes. Results Plasma metabolite profiles explained 14.1% of systolic BP variance and 10.6% of diastolic BP variance. These were attenuated to 3.1% and 1.4% respectively, when using residuals of BP after adjusting for age and sex. Top predictors for both systolic and diastolic BP included N-formylmethionine, several acylcarnitines and polyunsaturated fatty acids such as hexadecadienoate. These metabolites were significantly associated with higher systolic BP with estimates ranging from 3.0 to 4.5 mmHg per 1 SD increase in the adjusted models. Associations with hexadecadienoate, dihomolinoleate and catecholamine metabolites, including vanillactate had significant interactions (p<0.05) with ethnicity, and were only significant in subjects of non-African descent. Conclusions Plasma metabolome composition explained a large proportion of BP variance, but this association was attenuated when adjusting for confounders. Polyunsaturated fatty acids and catecholamine metabolites were only associated with BP in the non-African descent subjects. N-formylmethionine was the most consistent predictor for systolic BP across all subgroups. Future studies could focus on translating these findings in vitro in order to decipher the role of N-formylmethionine in BP regulation. FUNDunding Acknowledgement Type of funding sources: Public grant(s) – EU funding. Main funding source(s): Dutch Heart Foundation, the Netherlands Organization for Health Research and Development, the European Integration Fund and the European Union (Seventh Framework Programme) Explained variances of machine learning Linear regression models

Abstract WMP14: Association of Systolic Blood Pressure, LDL Cholesterol, and Hgb A1c With White Matter Hyperintensity on MRI in the ACCORDION MIND Study

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Adam H de Havenon ◽  
Tanya Turan ◽  
Rebecca Gottesman ◽  
Sharon Yeatts ◽  
Shyam Prabhakaran ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Risk Factor ◽  
Regression Models ◽  
Ldl Cholesterol ◽  
Vascular Risk Factor ◽  
White Matter Hyperintensity ◽  
Vascular Risk ◽  
Linear Regression Models ◽  
Risk Factor Profile ◽  
The Mean

Introduction: While retrospective studies have shown that poor control of vascular risk factors is associated with progression of white matter hyperintensity (WMH), it has not been studied prospectively. Hypothesis: We hypothesize that higher systolic blood pressure (SBP) mean, LDL cholesterol, and Hgb A1c will be correlated with WMH progression in diabetics. Methods: This is a secondary analysis of the Memory in Diabetes (MIND) substudy of the Action to Control Cardiovascular Risk in Diabetes Follow-on Study (ACCORDION). The primary outcome was WMH progression, evaluated by fitting linear regression models to the WMH volume on the month 80 MRI and adjusting for the WMH volume on the baseline MRI. The primary predictors were the mean values of SBP, LDL, and A1c from baseline to month 80. We defined a good vascular risk factor profile as mean SBP <120 mm Hg and mean LDL <120 mg/dL. Results: We included 292 patients, with a mean (SD) age of 62.6 (5.3) years and 55.8% male. The mean number of SBP, LDL, and A1c measurements per patient was 17, 5, and 12. We identified 86 (29.4%) patients with good vascular risk factor profile. In the linear regression models, mean SBP and LDL were associated with WMH progression and in a second fully adjusted model they both remained associated with WMH progression (Table). Those with a good vascular risk factor profile had less WMH progression (β Coefficient -0.80, 95% CI -1.42, -0.18, p=0.012). Conclusions: Our data reinforce prior research showing that higher SBP and LDL is associated with progression of WMH in diabetics, likely secondary to chronic microvascular ischemia, and suggest that control of these factors may have protective effects. This study has unique strengths, including prospective serial measurement of the exposures, validated algorithmic measurement methodology for WMH, and rigorous adjudication of study data. Clinical trials are needed to investigate the effect of vascular risk factor reduction on WMH progression.

scholarly journals Application of Machine Learning Techniques to Predict Mechanical Properties for Polyamide 2200 (PA12) in Additive Manufacturing

2019 ◽  
Author(s):  
Ivanna Baturynska
Keyword(s):  
Machine Learning ◽  
Mechanical Properties ◽  
Additive Manufacturing ◽  
Linear Regression ◽  
Prediction Accuracy ◽  
Regression Models ◽  
Tensile Modulus ◽  
Machine Learning Techniques ◽  
Linear Regression Models ◽  
Elongation At Break

Additive manufacturing (AM) is an attractive technology for manufacturing industry due to flexibility in design and functionality, but inconsistency in quality is one of the major limitations that does not allow utilizing this technology for production of end-use parts. Prediction of mechanical properties can be one of the possible ways to improve the repeatability of the results. The part placement, part orientation, and STL model properties (number of mesh triangles, surface, and volume) are used to predict tensile modulus, nominal stress and elongation at break for polyamide 2200 (also known as PA12). EOS P395 polymer powder bed fusion system was used to fabricate 217 specimens in two identical builds (434 specimens in total). Prediction is performed for XYZ, XZY, ZYX, and Angle orientations separately, and all orientations together. The different non-linear models based on machine learning methods have higher prediction accuracy compared with linear regression models. Linear regression models have prediction accuracy higher than 80% only for Tensile Modulus and Elongation at break in Angle orientation. Since orientation-based modeling has low prediction accuracy due to a small number of data points and lack of information about material properties, these models need to be improved in the future based on additional experimental work.

scholarly journals Application of Machine Learning Techniques to Predict the Mechanical Properties of Polyamide 2200 (PA12) in Additive Manufacturing

2019 ◽  
Vol 9 (6) ◽  
pp. 1060
Author(s):  
Ivanna Baturynska
Keyword(s):  
Machine Learning ◽  
Mechanical Properties ◽  
Additive Manufacturing ◽  
Linear Regression ◽  
Prediction Accuracy ◽  
Regression Models ◽  
Tensile Modulus ◽  
Machine Learning Techniques ◽  
Linear Regression Models ◽  
Elongation At Break

Additive manufacturing (AM) is an attractive technology for the manufacturing industry due to flexibility in its design and functionality, but inconsistency in quality is one of the major limitations preventing utilizing this technology for the production of end-use parts. The prediction of mechanical properties can be one of the possible ways to improve the repeatability of results. The part placement, part orientation, and STL model properties (number of mesh triangles, surface, and volume) are used to predict tensile modulus, nominal stress, and elongation at break for polyamide 2200 (also known as PA12). An EOS P395 polymer powder bed fusion system was used to fabricate 217 specimens in two identical builds (434 specimens in total). Prediction is performed for XYZ, XZY, ZYX, and Angle orientations separately, and all orientations together. The different non-linear models based on machine learning methods have higher prediction accuracy compared with linear regression models. Linear regression models only have prediction accuracy higher than 80% for Tensile Modulus and Elongation at break in Angle orientation. Since orientation-based modeling has low prediction accuracy due to a small number of data points and lack of information about the material properties, these models need to be improved in the future based on additional experimental work.

scholarly journals Predictive performance of linear regression models in estimation of Artemia salina abundance using field and remote sensing data

2021 ◽  
pp. 88-95
Author(s):  
D. Krivoguz ◽  
◽  
R. Borovskaya ◽  
Keyword(s):  
Machine Learning ◽  
Linear Regression ◽  
Spatial Data ◽  
Regression Models ◽  
Remote Sensing Data ◽  
Artemia Salina ◽  
Machine Learning Algorithms ◽  
Linear Regression Models ◽  
Light Spectrum ◽  
Fourth Degree

This research has been aimed at finding the possibilities for application of the linear regression models, as a part of the machine learning methods, in visual representation of the spatial patterns of Artemia salina distribution in the Southern Sivash. Development of such models allows for estimation of A. salina biomass in water bodies with high accuracy. For investigation of maximum absorption levels in different parts of the light spectrum, spectral signatures at all the monitoring stations have been compared with the satellite data, and the analysis of the absorption spectra for astaxanthin and hemoglobin has been conducted with a spectrophotometer. As a result, Sentinel-2 satellite looks very promising as a key spatial data provider that can be of major help in increasing the frequency of A. salina monitoring in the Southern Sivash. The linear regression models, fitted by the third and the fourth degree polynomials, have shown satisfactory results, suitable for their subsequent use in fisheries. On the other hand, it should be noted that these models are slightly prone to overfitting, which to some extent can distort further forecasts feeding upon the new data. In turn, linear regression models fitted by a polynomial of the first degree show less accurate results, but their advantages include the lack of tendency to overfit. It is also worth noting that small-sized datasets within the scope of this investigation do not appear to be problematic, and simple machine learning algorithms can provide good accuracy results, which are suitable for further application in this field.

Abstract P082: Lower Adiponectin Levels May Underlie Association Between Depressive Symptoms And Markers of Cardiometabolic Health

Circulation ◽  
2012 ◽  
Vol 125 (suppl_10) ◽  
Author(s):  
Cari J Clark ◽  
Qi Wang ◽  
Hongfei Guo ◽  
Joyce T Bromberger ◽  
Peter Mancuso ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Depressive Symptoms ◽  
Regression Models ◽  
Missing Values ◽  
Smoking Status ◽  
Depression Scale ◽  
Cardiometabolic Health ◽  
Enzyme Linked Immunosorbent Assay ◽  
Linear Regression Models ◽  
Cvd Risk

Introduction: Depressive symptoms have been linked to CVD risk factors, including metabolic dysregulation. One pathway by which depression may influence CVD risk is via alterations in adiponectin, an abundant adipocytokine with anti-inflammatory effects. This mechanism has not been studied in population-based samples. Hypothesis: The relationship of depressive symptoms with metabolic syndrome (MetSyn) and Framingham Risk Score (FRS) will be partly mediated by adiponectin. Methods: Participants were 581 women (61.3% white; 38.7% black) from the Chicago and Pittsburgh sites of the Study of Women’s Health Across the Nation. Adiponectin was measured from stored serum specimens and assayed in duplicate using a commercially available enzyme linked immunosorbent assay and log transformed for analysis. Depressive symptoms were measured with the 20-item Center for Epidemiological Studies Depression Scale (CES-D); a standard cutoff (>16) was used to determine clinically significant symptoms. MetSyn was defined by ATP-III criteria and considered present if the participant had at least 3 of the following: waist circumference >88cm; triglycerides >150 mg/dl; HDL cholesterol < 50 mg/dl; blood pressure > 130 mmHg systolic and / or 85 mmHg diastolic; impaired fasting glucose (>110 mg/dl) or diabetes. The FRS was defined by the participant’s age, smoking status, blood pressure, cholesterol, and use of anti-hypertensives. Logistic regression models were constructed to examine the cross-sectional relationship between depressive symptoms and MetSyn controlling for age, race and study site. A subsequent model included adiponectin to evaluate whether it attenuated the observed association. Linear regression models were used to conduct the same analysis with FRS as the outcome. Due to missing values, analytic sample sizes were 558 for MetSyn and 568 for FRS. Results: 147 women (25.3%) had elevated CES-D scores and 113 (20.7%) met criteria for MetSyn. Average FRS was 8.7 (sd=4.6) and the mean, untransformed adiponectin value was 9.9 (sd=4.9) μ g/mL. In models adjusted for age, race, and study site, women with high CES-D scores had increased odds of MetSyn (OR=1.64; 95% CI=1.03, 2.60) and a higher FRS (estimate=0.98; se=0.41, p<.02). Separate bivariate analyses showed that adiponectin was inversely related to CES-D scores (p=.03), MetSyn (p<.001) and FRS (p<.001). Subsequently including adiponectin in the regression models attenuated the associations between CES-D and MetSyn (OR=1.45; 95% CI=0.89, 2.36) and FRS (estimate=0.76; se=0.41; p=.06). Conclusions: Adiponectin may partially explain the relation between depressive symptoms and measures of cardiometabolic health. Longitudinal studies are needed to more fully understand the temporality of these associations. Supported by NIH/DHHS grants HL091290, AG012505, AG012546, MH59770, AG17719.

scholarly journals May Measurement Month 2018: an analysis of blood pressure screening results from China

2020 ◽  
Vol 22 (Supplement_H) ◽  
pp. H40-H42
Author(s):  
Xin Chen ◽  
Yan Li ◽  
Zhe Hu ◽  
Min Liu ◽  
Jing Yu ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Regression Models ◽  
Alcohol Intake ◽  
Nationwide Survey ◽  
Overweight And Obesity ◽  
Hypertension Management ◽  
Linear Regression Models ◽  
Study Results ◽  
Central Database ◽  
And Control

Abstract To further improve awareness, treatment, and control of hypertension, the May Measurement Month (MMM) campaign continued in 2018 in China. Study subjects were adults aged 18 years or more, ideally those who had not their blood pressure (BP) measured for at least a year. Blood pressure was measured three times consecutively with a 1-min interval in the sitting position, using automated BP monitors in 288 342 participants and transmitted to a central database by a smartphone app. Questionnaire data were collected with the same app. After imputation, the overall proportion of hypertension was 29.8%. Of those with hypertension, the rates of awareness, treatment, and control were 62.3%, 57.3%, and 35.9%, respectively. In analysis based on linear regression models, both systolic and diastolic BP were higher with cigarette smoking, alcohol intake, and overweight and obesity. Our study results suggest that hypertension management is improving in comparison with the data in MMM 2017 and the nationwide survey in 2012–15, and several known lifestyle factors are key to hypertension management.

scholarly journals Dialysis adequacy predictions using a machine learning method

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Hyung Woo Kim ◽  
Seok-Jae Heo ◽  
Jae Young Kim ◽  
Annie Kim ◽  
Chung-Mo Nam ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Regression Models ◽  
Measured Data ◽  
Chronic Hemodialysis ◽  
Linear Regression Models ◽  
Learning Models ◽  
Dialysis Adequacy ◽  
Gated Recurrent Unit ◽  
Machine Learning Models

AbstractDialysis adequacy is an important survival indicator in patients with chronic hemodialysis. However, there are inconveniences and disadvantages to measuring dialysis adequacy by blood samples. This study used machine learning models to predict dialysis adequacy in chronic hemodialysis patients using repeatedly measured data during hemodialysis. This study included 1333 hemodialysis sessions corresponding to the monthly examination dates of 61 patients. Patient demographics and clinical parameters were continuously measured from the hemodialysis machine; 240 measurements were collected from each hemodialysis session. Machine learning models (random forest and extreme gradient boosting [XGBoost]) and deep learning models (convolutional neural network and gated recurrent unit) were compared with multivariable linear regression models. The mean absolute percentage error (MAPE), root mean square error (RMSE), and Spearman’s rank correlation coefficient (Corr) for each model using fivefold cross-validation were calculated as performance measurements. The XGBoost model had the best performance among all methods (MAPE = 2.500; RMSE = 2.906; Corr = 0.873). The deep learning models with convolutional neural network (MAPE = 2.835; RMSE = 3.125; Corr = 0.833) and gated recurrent unit (MAPE = 2.974; RMSE = 3.230; Corr = 0.824) had similar performances. The linear regression models had the lowest performance (MAPE = 3.284; RMSE = 3.586; Corr = 0.770) compared with other models. Machine learning methods can accurately infer hemodialysis adequacy using continuously measured data from hemodialysis machines.

scholarly journals Identification of Diagnostic Plasma Biomarkers of Coronary Microvascular Disease in Postmenopausal Women Using Machine Learning Methods

2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. A288-A288
Author(s):  
Alicia Arredondo Eve ◽  
Elif Tunc ◽  
Yu-Jeh Liu ◽  
Saumya Agrawal ◽  
Huriye Huriye Erbak Yilmaz ◽  
...  
Keyword(s):  
Machine Learning ◽  
Postmenopausal Women ◽  
Microvascular Disease ◽  
Classification Performance ◽  
Learning Approaches ◽  
Plasma Samples ◽  
Coronary Syndrome ◽  
Healthy Women ◽  
Metabolite Profiles ◽  
Plasma Metabolite

Abstract Introduction: Coronary microvascular disease (CMD) affects small arteries that feed the heart and is more prevalent in postmenopausal women. Since CMD and Coronary artery disease (CAD) have distinct pathologies, but are treated the same way, the majority of the patients with CMD do not receive a proper diagnosis and treatment, which in turn results in higher rates of adverse future events such as heart failure, sudden cardiac death, and acute coronary syndrome (ACS). Previously, we performed full metabolite profiling of plasma samples using GC-MS analysis and tested their classification performance using machine learning approaches. This initial proof-of-concept study showed that plasma metabolite profiles can be used to develop diagnostic signatures for CMD. In the current study, we hypothesize that plasma metabolite and protein composition is different for postmenopausal women with no heart disease, with CAD, or with CMD. Methods: We obtained plasma samples from 70 postmenopausal women who are healthy, women who have CMD, and women who have CAD at the time of blood collection. In addition to GC-MS metabolite profiles, we performed LC-MS metabolomic profiling, and proteomic profiling of a panel of 92 proteins that were implicated in cardiometabolic disease. We identified a combination of metabolites and proteins, and further tested their classification performance using machine learning approaches to identify potential circulating biomarkers for CMD. Results: We identified a comprehensive list of metabolites and proteins that were involved in endothelial cell function, nitric oxide metabolism and inflammation, which significantly different in plasma from women with CMD. We further validated difference in the level of several protein biomarkers, such as RAGE, PTX3, AGRP, CNTN1, and MMP-3, which are statistically significantly higher in postmenopausal women with CMD when compared with healthy women or women with CAD. Conclusion: Our research identified a group of potential molecules that can be used in the design of easy and low-cost blood biomarkers for the clinical diagnosis of CMD.

scholarly journals Fasting plasma glucose, lipid ratios, and atherogenic coefficient are the risk factors for hypertension in chronic kidney disease patients on hemodialysis: A report from the Regional High Speciality Hospital of Peninsular Yucatan, Mexico

2020 ◽  
Vol 83 (3) ◽  
pp. 251-260
Author(s):  
Sudip Datta Banik ◽  
Ricardo Hernández Cardoza ◽  
Rosa María Méndez González ◽  
Ana Ligia Gutiérrez Solis
Keyword(s):  
Risk Factors ◽  
Blood Pressure ◽  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Linear Regression ◽  
Lipid Profile ◽  
Plasma Glucose ◽  
Regression Models ◽  
Linear Regression Models ◽  
Multiple Linear Regression Models

AbstractChronic kidney disease (CKD) is associated with the development of cardivascular disease (CVD). CKD is one of the major public health problems in Mexico. Derived parameters of lipid profile, namely atherogenic index of plasma (AIP), atherogenic coefficient (AC), and Castelli risk index (CRI I and CRI II) are useful for predicting hypertension among CKD patients on hemodialysis that are not widely reported from Mexico. Objective of the present study was to find interrelationships among blood pressure, fasting plasma glucose (FPG), and derived parameters of lipid profile (AIP, AC, CRI-I, and CRI-II) among adult CKD patients on hemodialysis in a hospital in Yucatan, Mexico. Methods: An observational study was performed using the medical records (2016 and 2017) of 47 CKD patients on hemodialysis in the Regional High Speciality Hospital of Yucatan Peninsula (HRAEPY in Spanish acronym). Multiple linear regression models were developed to evaluate the use of FPG level and derived parameters of lipid profile (AC, CRI-I, and CRI-II) as risk factors predicting mean arterial pressure (MAP). Results showed remarkable prevalence of excess weight (55% overweight, 15% obesity) and hypertension (64%) in the sample. Correlation coeffcients and multiple linear regression models showed significant rise of blood pressure in association with elevated FPG level and derived lipid profile parameters. The results confirm the use of FPG, AC, CRI-I and CRI-II as the indicators for an early diagnosis of hypertension and related CVDs among CKD patients on hemodialysis.

Sign in / Sign up

Export Citation Format

Share Document