scholarly journals Predicting stroke risk in Chinese hypertensive population using machine learning

2021 ◽  
Vol 42 (Supplement_1) ◽  
Author(s):  
X Huang ◽  
T Y Cao ◽  
Y P Wei ◽  
B Xu ◽  
H Y Wu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
General Framework ◽  
Primary Outcome ◽  
Stroke Risk ◽  
Prediction Method ◽  
Analysis Method ◽  
Hypertensive Population ◽  
The Mean ◽  
Very High

Abstract Background Stroke is the leading cause of death in China, and the stroke burden is especially high in rural areas. Risk prediction is essential for primary prevention of stroke. However, uncertainty remains about the optimal methodology for analyzing stroke risk. In this study, we aim to determine the most effective stroke prediction method in a targeted population and establish a general framework and pipeline for future analysis. Purpose 1) to determine the most effective stroke prediction method in a targeted population and 2) to establish a general framework and pipeline for future analysis. Methods Data were obtained from the China Stroke Primary Prevention Trial (CSPPT), a randomized, double-blind, multi-center clinical trial. 20,702 hypertensive patients without prior history of stroke were included in the study. The primary outcome was new nonfatal and fatal stroke (ischemic or hemorrhagic) occurring between baseline and follow-up (a median of 4.2 years). All suspected stroke cases were collected and further validated by the event adjudication committee. We compared two regression models (logistic regression and step wise logistic regression) and two machine learning methods (extreme gradient boosting and random forest). All models were trained using questionnaire data with and without laboratory data, then analyzed and compared. The primary outcome was defined as first stroke. Accuracy, sensitivity, specificity and AUCs (area under receiver operating characteristic curve) were used to assess each model. AUCs were used to evaluate the performance of each analysis method. Results In our data set with 20,702 samples and 127 variables, the highest AUCs (0.775 (0.725–0.826)) were observed with RUS (random under sampling) applied to RF (random forest). Before applying data balancing techniques, all analysis methods showed very low sensitivity (around 0.01), very high accuracy (around 0.97), and very high specificity (around 1.00). The mean AUCs were 0.741 (0.678–0.803). After data balancing techniques were applied, we observed an increase in sensitivity and decreases in accuracy and specificity. Different data balancing techniques had different effects on analysis methods. No significant effect on AUCs was observed; the range of increase and decrease was around 0.01. Similar overall patterns were observed when training with laboratory test data added. The mean AUCs were 0.739 (0.679–0.799) and 0.734 (0.674–9.795) for all models using data with and without laboratory test respectively. The 10 most important variables as determined by the model were selected as stroke risk predictors for all analysis models. Conclusion The most effective stroke prediction method in this Chinese rural hypertensive population is RUS applied to RF. The optimal analysis method and variable selection depends on data-specific features. FUNDunding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): Key R&D Projects, Jiangxi [20203BBGL73173] National Key Research and Development Program [2016YFE0205400]

Abstract 133: White Matter Hyperintensity and Cardiovascular Disease Outcomes: A Secondary Analysis of the SPRINT MIND Trial

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Adam H de Havenon ◽  
Tanya Turan ◽  
Sharon Yeatts ◽  
Rebecca Gottesman ◽  
Shyam Prabhakaran ◽  
...  
Keyword(s):  
White Matter ◽  
Primary Outcome ◽  
Stroke Risk ◽  
Secondary Analysis ◽  
White Matter Hyperintensity ◽  
Secondary Outcome ◽  
Cox Models ◽  
Disease Outcomes ◽  
Hazard Ratios ◽  
The Mean

Background: The Systolic Blood Pressure Intervention Trial (SPRINT) randomized patients to a goal SBP <120 mm Hg vs. <140 mm Hg . A subset of patients enrolled in SPRINT MIND, which performed a baseline MRI and measured white matter hyperintensity volume (WMHv). We evaluated the association between WMHv and cardiovascular events. Methods: The primary outcome was a composite of stroke, MI, ACS, decompensated CHF, or CVD death. The secondary outcome was stroke. The WMHv was divided into quartiles. We fit Cox models to the outcomes and report adjusted hazard ratios for the quartiles of WMHv, and stratified by SPRINT treatment arm. Results: Among 719 included patients, the mean WMHv in the quartiles was 0.34, 1.09, 2.61, and 10.8 mL. The primary outcome occurred in 51/719 (7.1%) and the secondary outcome in 10/719 (1.4%). The WMHv was associated with both outcomes (Table 1, Figure 1). After stratifying by treatment arm, we found the association persisted in the standard, but not intensive, treatment arm (Table 2). However, the interaction term between WMHv and treatment arm was not significant. Conclusions: We observed that degree of WMH was associated with CVD and stroke risk in SPRINT MIND. The risk may be attenuated in patients randomized to intensive BP lowering. Trials are needed to determine if intensive BP lowering can prospectively reduce the high cardiovascular risk in patients with WMH.

Predicting Stroke Risk in a Chinese Hypertensive Population Using Machine Learning

2021 ◽  
Author(s):  
Xiao Huang ◽  
Tianyu Cao ◽  
Liangziqian Chen ◽  
Hanyu Wu ◽  
Junpei Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Stroke Risk ◽  
Hypertensive Population

Seismic Reliability of UHV Porcelain arrester with Damping System

2014 ◽  
Vol 8 (1) ◽  
pp. 222-228 ◽  
Author(s):  
Xuesong Zhang ◽  
Meigen Cao ◽  
Zhicheng Lu
Keyword(s):  
Fragility Curves ◽  
Single Point ◽  
Analysis Method ◽  
Element Analysis ◽  
Seismic Reliability ◽  
Seismic Motion ◽  
Earthquake Model ◽  
New Type ◽  
The Mean ◽  
Very High

The vulnerability of the UHV porcelain arrester is very high under strong earthquake. To rise seismic reliability of the UHV porcelain arrester, a new type lead damper, which is a new patent product, is installed at the bottom of the equipment. To investigate the damping effect of the dampers, the experimental research and finite element analysis on seismic behavior of the UHV porcelain arrester with and without the dampers are carried out by means of single point input and single point output(SISO)measurement. The analyzed results show that the damper is well functioning, effectively decreasing stochastic earthquake response of the UHV porcelain arrester, thus the seismic reliability of porcelain arrester is improved. A seismic reliability analysis method is put forward based on the stochastic earthquake model .The mean and standard deviation of the seismic responses of the arrester with and without the dampers under the different site conditions are gained based on physical stochastic seismic motion model. Then its seismic reliability is calculated by the FOSM, and the fragility curves of the arrester are built. Calculation result shows that seismic reliability of the arrester with the dampers can be effectively enhanced under Ms 8.0 earthquake. A conclusion is given that the damper is capable to improve the seismic reliability of the UHV porcelain arrester effectively, and may be widely applied to the seismic design of the UHV porcelain arrester.

scholarly journals A Simple and Interpretable Severe Intraventricular Hemorrhage Prediction Model for Extremely Low Birth Weight Infants Using Machine Learning

2021 ◽  
Author(s):  
Felipe Yu Matsushita ◽  
Vera Lúcia Jornada Krebs ◽  
Werther Brunow de Carvalho
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Birth Weight ◽  
Low Birth Weight ◽  
Intraventricular Hemorrhage ◽  
Extremely Low Birth Weight ◽  
Single Center ◽  
Tree Model ◽  
Low Birth Weight Infants ◽  
The Mean

Abstract Severe intraventricular hemorrhage (sIVH) is a catastrophic event with serious neurocognitive impairment in preterm infants. Because sIVH is a complex multifactorial disease, determining which patients require special attention to prevent sIVH is challenging. This study aimed to evaluate an easy interpretable decision-tree model to identify extremely preterm infants with a higher risk of severe intraventricular hemorrhage. All infants admitted to a single-center tertiary intensive care unit in São Paulo, Brazil, from 2012 to 2017, with a birth weight less than 1000 grams and at least one cranial ultrasound after three days of life were included. The association of risk factors with sIVH was assessed using logistic regression. Univariate analysis, stepwise logistic regression, correlation matrix, Boruta, and XGBoost were used to select features. In this single-center, retrospective cohort of 190 extremely low birth weight infants, the mean gestational age was 27.5 (2.2) weeks and the mean birth weight was 748 (161) grams. A total of forty-two newborns (22.1%) developed severe intraventricular hemorrhage. Machine learning tools identified three features (pH, base excess, and gestational age) that predict severe intraventricular hemorrhage with an AUC of 0.857. Low pH levels appear to be a key factor in identifying the great majority of cases that require additional attention. Conclusions: We suggest a simple and interpretable decision-tree model to promptly identify extremely low birth weight infants at the highest risk of severe intraventricular hemorrhage.

Monoamine Oxidase and Other Mitochondrial Enzymes in Density Subpopulations of Human Platelets

1988 ◽  
Vol 59 (01) ◽  
pp. 029-033 ◽  
Author(s):  
K G Chamberlain ◽  
D G Penington
Keyword(s):  
Monoamine Oxidase ◽  
Protein Concentration ◽  
Close Correlation ◽  
Human Platelets ◽  
Mitochondrial Enzymes ◽  
Density Fractions ◽  
Percoll Gradients ◽  
Normal Human ◽  
The Mean ◽  
Very High

SummaryNormal human platelets have been separated according to density on continuous Percoll gradients and the platelet distribution divided into five fractions containing approximately equal numbers of platelets. The mean volumes and protein contents of the platelets in each fraction were found to correlate positively with density while the protein concentration did not differ significantly between the fractions. Four mitochondrial enzymes (monoamine oxidase, glutamate dehydrogenase, cytochrome oxidase and NADP-dependent isocitrate dehydrogenase) were assayed and their activities per unit volume were found to increase in a very similar monotonie fashion with platelet density. When MAO and GDH were assayed on the same set of density fractions the correlation between the two activities was very high (r = 0.94–1.00, p <0.001) and a similar close correlation was found between MAO and ICDH. The results support the hypothesis that high density platelets either have a higher concentration of mitochondria or have larger mitochondria than low density platelets.

Voronoi Analysis of a Soccer Game

2004 ◽  
Vol 9 (3) ◽  
pp. 233-240 ◽  
Author(s):  
S. Kim
Keyword(s):  
Collective Behavior ◽  
Quantitative Assessment ◽  
The Other ◽  
Analysis Method ◽  
Random System ◽  
Soccer Game ◽  
Voronoi Polygons ◽  
The Mean

This paper describes a Voronoi analysis method to analyze a soccer game. It is important for us to know the quantitative assessment of contribution done by a player or a team in the game as an individual or collective behavior. The mean numbers of vertices are reported to be 5–6, which is a little less than those of a perfect random system. Voronoi polygons areas can be used in evaluating the dominance of a team over the other. By introducing an excess Voronoi area, we can draw some fruitful results to appraise a player or a team rather quantitatively.

Predictors of remission from body dysmorphic disorder after internet-delivered cognitive behavior therapy: a machine learning approach

2019 ◽  
Author(s):  
Oskar Flygare ◽  
Jesper Enander ◽  
Erik Andersson ◽  
Brjánn Ljótsson ◽  
Volen Z Ivanov ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forests ◽  
Clinical Utility ◽  
Body Dysmorphic Disorder ◽  
Prediction Models ◽  
Behavioral Therapy ◽  
Learning Approach ◽  
Learning Approaches ◽  
Machine Learning Approach

**Background:** Previous attempts to identify predictors of treatment outcomes in body dysmorphic disorder (BDD) have yielded inconsistent findings. One way to increase precision and clinical utility could be to use machine learning methods, which can incorporate multiple non-linear associations in prediction models. **Methods:** This study used a random forests machine learning approach to test if it is possible to reliably predict remission from BDD in a sample of 88 individuals that had received internet-delivered cognitive behavioral therapy for BDD. The random forest models were compared to traditional logistic regression analyses. **Results:** Random forests correctly identified 78% of participants as remitters or non-remitters at post-treatment. The accuracy of prediction was lower in subsequent follow-ups (68%, 66% and 61% correctly classified at 3-, 12- and 24-month follow-ups, respectively). Depressive symptoms, treatment credibility, working alliance, and initial severity of BDD were among the most important predictors at the beginning of treatment. By contrast, the logistic regression models did not identify consistent and strong predictors of remission from BDD. **Conclusions:** The results provide initial support for the clinical utility of machine learning approaches in the prediction of outcomes of patients with BDD. **Trial registration:** ClinicalTrials.gov ID: NCT02010619.

Der Sommer und Herbst 2003 aus phänologischer Sicht | Summer and autumn 2003 from a phenological point of view

2004 ◽  
Vol 155 (5) ◽  
pp. 142-145 ◽  
Author(s):  
Claudio Defila
Keyword(s):  
Time Series ◽  
New Record ◽  
Point Of View ◽  
Late Spring ◽  
Mean Deviation ◽  
Leaf Shedding ◽  
The Mean ◽  
Very High ◽  
Selection Of

The record-breaking heatwave of 2003 also had an impact on the vegetation in Switzerland. To examine its influences seven phenological late spring and summer phases were evaluated together with six phases in the autumn from a selection of stations. 30% of the 122 chosen phenological time series in late spring and summer phases set a new record (earliest arrival). The proportion of very early arrivals is very high and the mean deviation from the norm is between 10 and 20 days. The situation was less extreme in autumn, where 20% of the 103 time series chosen set a new record. The majority of the phenological arrivals were found in the class «normal» but the class«very early» is still well represented. The mean precocity lies between five and twenty days. As far as the leaf shedding of the beech is concerned, there was even a slight delay of around six days. The evaluation serves to show that the heatwave of 2003 strongly influenced the phenological events of summer and spring.

AN EFFICIENT MACHINE LEARNING MODEL FOR PREDICTION OF ACUTE MYOCARDIAL INFARCTION

2020 ◽  
Vol 13 ◽  
Author(s):  
Dhilsath Fathima.M ◽  
S. Justin Samuel ◽  
R. Hari Haran
Keyword(s):  
Machine Learning ◽  
Myocardial Infarction ◽  
Acute Myocardial Infarction ◽  
Logistic Regression ◽  
Decision Tree ◽  
Learning Model ◽  
Training Dataset ◽  
Data Set ◽  
Machine Learning Model ◽  
Proposed Model

Aim: This proposed work is used to develop an improved and robust machine learning model for predicting Myocardial Infarction (MI) could have substantial clinical impact. Objectives: This paper explains how to build machine learning based computer-aided analysis system for an early and accurate prediction of Myocardial Infarction (MI) which utilizes framingham heart study dataset for validation and evaluation. This proposed computer-aided analysis model will support medical professionals to predict myocardial infarction proficiently. Methods: The proposed model utilize the mean imputation to remove the missing values from the data set, then applied principal component analysis to extract the optimal features from the data set to enhance the performance of the classifiers. After PCA, the reduced features are partitioned into training dataset and testing dataset where 70% of the training dataset are given as an input to the four well-liked classifiers as support vector machine, k-nearest neighbor, logistic regression and decision tree to train the classifiers and 30% of test dataset is used to evaluate an output of machine learning model using performance metrics as confusion matrix, classifier accuracy, precision, sensitivity, F1-score, AUC-ROC curve. Results: Output of the classifiers are evaluated using performance measures and we observed that logistic regression provides high accuracy than K-NN, SVM, decision tree classifiers and PCA performs sound as a good feature extraction method to enhance the performance of proposed model. From these analyses, we conclude that logistic regression having good mean accuracy level and standard deviation accuracy compared with the other three algorithms. AUC-ROC curve of the proposed classifiers is analyzed from the output figure.4, figure.5 that logistic regression exhibits good AUC-ROC score, i.e. around 70% compared to k-NN and decision tree algorithm. Conclusion: From the result analysis, we infer that this proposed machine learning model will act as an optimal decision making system to predict the acute myocardial infarction at an early stage than an existing machine learning based prediction models and it is capable to predict the presence of an acute myocardial Infarction with human using the heart disease risk factors, in order to decide when to start lifestyle modification and medical treatment to prevent the heart disease.

Sign in / Sign up

Export Citation Format

Share Document