Machine learning algorithms identify demographics, dietary features, and blood biomarkers associated with stroke records

Liu, Jundong; Chou, Elizabeth L.; Lau, Kui Kai; Woo, Peter Yat Ming; Li, Jun; Chan, Kei Hang Katie

doi:10.1016/j.jns.2022.120335

Cited by 7 publications

(19 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Examining the general characteristics of our study population, we observed that individuals diagnosed with CVD were predominantly older, male, had a higher BMI, and showed elevated levels of several clinical parameters, such as systolic blood pressure, HDL, and uric acid. These observations largely align with the known demographic and physiological risk factors of CVD [ 8 , 9 , 10 ]. Glycated hemoglobin, uric acid, and

, which are markers of diabetes, gout, and kidney disease, respectively, have previously been associated with CVD [ 15 , 16 , 17 ].…”

Section: Discussionsupporting

confidence: 70%

“…Although direct comparison with previous studies presents challenges due to differences in variable selection, the majority of the risk factors listed in our chart align with findings from prior research. Age, systolic and diastolic blood pressure, and current smoker status have consistently been highlighted as significant variables in CVD classification or risk of death [ 8 , 24 , 26 ]. Glycated hemoglobin was also identified as a risk factor for CVD onset using ML methodologies [ 27 ].…”

Section: Discussionmentioning

confidence: 99%

“…Fortunately, comprehensive databases such as the National Health and Nutrition Examination Survey (NHANES) allow for detailed nutritional profiling. By employing advanced ML techniques on such extensive data sets, it becomes feasible to discern complex interactions between different types of nutrients and their nonlinear relationships with CVD outcomes—far beyond the scope of traditional regression models or composite nutritional indices [ 8 , 9 , 10 , 11 ]. However, there’s still uncertainty regarding whether nutritional data from dietary recalls can significantly enhance CVD mortality risk prediction beyond that achieved using traditionally used biomarkers, such as blood pressure or cholesterol levels.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables

Martin-Morales,

Yamamoto,

Inoue

et al. 2023

Nutrients

View full text Add to dashboard Cite

Cardiovascular disease (CVD) is one of the primary causes of death around the world. This study aimed to identify risk factors associated with CVD mortality using data from the National Health and Nutrition Examination Survey (NHANES). We created three models focusing on dietary data, non-diet-related health data, and a combination of both. Machine learning (ML) models, particularly the random forest algorithm, demonstrated robust consistency across health, nutrition, and mixed categories in predicting death from CVD. Shapley additive explanation (SHAP) values showed age, systolic blood pressure, and several other health factors as crucial variables, while fiber, calcium, and vitamin E, among others, were significant nutritional variables. Our research emphasizes the importance of comprehensive health evaluation and dietary intake in predicting CVD mortality. The inclusion of nutrition variables improved the performance of our models, underscoring the utility of dietary intake in ML-based data analysis. Further investigation using large datasets with recurring dietary recalls is necessary to enhance the effectiveness and interpretability of such models.

show abstract

, which are markers of diabetes, gout, and kidney disease, respectively, have previously been associated with CVD [ 15 , 16 , 17 ].…”

Section: Discussionsupporting

confidence: 70%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables

Martin-Morales,

Yamamoto,

Inoue

et al. 2023

Nutrients

View full text Add to dashboard Cite

show abstract

“…Nomograms are helpful for predicting the probability of suffering from disease ( Park, 2018 ). Previous studies have shown that nomograms are widely used in the prediction of stroke, including establishing prediction models to identify blood biomarkers or medical imaging data associated with stroke and stroke implications ( Liu J. et al, 2022 ; Liu L. et al, 2022 ; Yang et al, 2022 ). All these results indicated that the nomogram is stable and accurate in predicting the probability of suffering stroke, which was in line with our study.…”

Section: Discussionmentioning

confidence: 99%

Development, validation, and visualization of a novel nomogram to predict stroke risk in patients

Wang

et al. 2023

Front. Aging Neurosci.

View full text Add to dashboard Cite

BackgroundStroke is the second leading cause of death worldwide and a major cause of long-term neurological disability, imposing an enormous financial burden on families and society. This study aimed to identify the predictors in stroke patients and construct a nomogram prediction model based on these predictors.MethodsThis retrospective study included 11,435 participants aged >20 years who were selected from the NHANES 2011–2018. Randomly selected subjects (n = 8531; 75%) and the remaining subjects comprised the development and validation groups, respectively. The least absolute shrinkage and selection operator (LASSO) binomial and logistic regression models were used to select the optimal predictive variables. The stroke probability was calculated using a predictor-based nomogram. Nomogram performance was assessed by the area under the receiver operating characteristic curve (AUC) and the calibration curve with 1000 bootstrap resample validations. Decision curve analysis (DCA) was performed to evaluate the clinical utility of the nomogram.ResultsAccording to the minimum criteria of non-zero coefficients of Lasso and logistic regression screening, older age, lower education level, lower family income, hypertension, depression status, diabetes, heavy smoking, heavy drinking, trouble sleeping, congestive heart failure (CHF), coronary heart disease (CHD), angina pectoris and myocardial infarction were independently associated with a higher stroke risk. A nomogram model for stroke patient risk was established based on these predictors. The AUC (C statistic) of the nomogram was 0.843 (95% CI: 0.8186–0.8430) in the development group and 0.826 (95% CI: 0.7811, 0.8716) in the validation group. The calibration curves after 1000 bootstraps displayed a good fit between the actual and predicted probabilities in both the development and validation groups. DCA showed that the model in the development and validation groups had a net benefit when the risk thresholds were 0–0.2 and 0–0.25, respectively.DiscussionThis study effectively established a nomogram including demographic characteristics, vascular risk factors, emotional factors and lifestyle behaviors to predict stroke risk. This nomogram is helpful for screening high-risk stroke individuals and could assist physicians in making better treatment decisions to reduce stroke occurrence.

show abstract

“…Due to substantial missingness and outliers, data preprocessing was performed. We assigned “NA” to the data with “7, 9, 77, 99,” deleted the variables with more than 30% missing values ( 22 , 23 ), and used the MI package in the R software to perform multiple interpolation for the variables with less than 30% missing values. Summary statistics were calculated following imputation.…”

Section: Methodsmentioning

confidence: 99%

Application of machine learning algorithms to identify people with low bone density

Xu,

Chen,

Yao

et al. 2024

Front. Public Health

View full text Add to dashboard Cite

BackgroundOsteoporosis is becoming more common worldwide, imposing a substantial burden on individuals and society. The onset of osteoporosis is subtle, early detection is challenging, and population-wide screening is infeasible. Thus, there is a need to develop a method to identify those at high risk for osteoporosis.ObjectiveThis study aimed to develop a machine learning algorithm to effectively identify people with low bone density, using readily available demographic and blood biochemical data.MethodsUsing NHANES 2017–2020 data, participants over 50 years old with complete femoral neck BMD data were selected. This cohort was randomly divided into training (70%) and test (30%) sets. Lasso regression selected variables for inclusion in six machine learning models built on the training data: logistic regression (LR), support vector machine (SVM), gradient boosting machine (GBM), naive Bayes (NB), artificial neural network (ANN) and random forest (RF). NHANES data from the 2013–2014 cycle was used as an external validation set input into the models to verify their generalizability. Model discrimination was assessed via AUC, accuracy, sensitivity, specificity, precision and F1 score. Calibration curves evaluated goodness-of-fit. Decision curves determined clinical utility. The SHAP framework analyzed variable importance.ResultsA total of 3,545 participants were included in the internal validation set of this study, of whom 1870 had normal bone density and 1,675 had low bone density Lasso regression selected 19 variables. In the test set, AUC was 0.785 (LR), 0.780 (SVM), 0.775 (GBM), 0.729 (NB), 0.771 (ANN), and 0.768 (RF). The LR model has the best discrimination and a better calibration curve fit, the best clinical net benefit for the decision curve, and it also reflects good predictive power in the external validation dataset The top variables in the LR model were: age, BMI, gender, creatine phosphokinase, total cholesterol and alkaline phosphatase.ConclusionThe machine learning model demonstrated effective classification of low BMD using blood biomarkers. This could aid clinical decision making for osteoporosis prevention and management.

show abstract

Machine learning algorithms identify demographics, dietary features, and blood biomarkers associated with stroke records

Cited by 7 publications

References 55 publications

Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables

Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables

Development, validation, and visualization of a novel nomogram to predict stroke risk in patients

Application of machine learning algorithms to identify people with low bone density

Contact Info

Product

Resources

About