Modelling of longitudinal data to predict cardiovascular disease risk: a methodological review

Stevens, David P.; Lane, Deirdre A.; Harrison, Stephanie L; Lip, Gregory Y.H.; Kolamunnage‐Dona, Ruwanthi

doi:10.1186/s12874-021-01472-x

Cited by 7 publications

(4 citation statements)

References 117 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unhealthy lifestyle habits, including a poor and unbalanced diet, lack of physical exercise, obesity, excessive smoking, and drinking, can lead to CVD. Risk prediction models can be used to understand and manage CVD and have become an essential part of the clinical decision making [ 121 ]. Many risk prediction models for CVD use one data point per patient (usually at the baseline), such as the widely used Framingham risk score, which predicts the risk of coronary heart disease [ 122 ], or QRISK3, which predicts risk of CVD in a subset of the UK population and is widely used in CVD risk stratification in the UK [ 123 ].…”

Section: Role Of Tea Plant and Related Compounds In Cardiovascular Di...mentioning

confidence: 99%

Tea Plant (Camellia sinensis): A Current Update on Use in Diabetes, Obesity, and Cardiovascular Disease

et al. 2022

View full text Add to dashboard Cite

The tea plant (C. sinensis) has traditionally been consumed worldwide as “tea” for its many health benefits, with the potential for the prevention and therapy of various conditions. Regardless of its long history, the use of tea plants in modern times seems not to have changed much, as the beverage remains the most popular form. This review aimed to compile scientific information about the role and action of tea plants, as well as their status concerning clinical applications, based on the currently available evidence, with a focus on metabolic syndrome, mainly covering obesity, diabetes, and cardiovascular disease. It has been recognized that these diseases pose a significant threat to public health, and the development of effective treatment and prevention strategies is necessary but still challenging. In this article, the potential benefits of tea plants and their derived bioactive components (such as epigallocatechin-3-gallate) as anti-obesity, anti-diabetic, and anti-cardiovascular agents are clearly shown and emphasized, along with their mechanisms of action. However, according to the status of the clinical translation of tea plants, particularly in drug development, more substantial efforts in well-designed, randomized, controlled trials are required to expand their applications in treating the three major metabolic disorders and avoiding the toxicity caused by overconsumption.

show abstract

Section: Role Of Tea Plant and Related Compounds In Cardiovascular Di...mentioning

confidence: 99%

Tea Plant (Camellia sinensis): A Current Update on Use in Diabetes, Obesity, and Cardiovascular Disease

et al. 2022

View full text Add to dashboard Cite

show abstract

“…presence or absence of disease) or decision trees (e.g. high-risk or low-risk) [9,10]. However, these approaches are limited in their accuracy, as they tend to rely on a single set of features and metrics that can be used to accurately predict the risk of heart disease.…”

Section: Introductionmentioning

confidence: 99%

An Efficient Hybrid Classification Model for Heart Disease Prediction

Munsif,

Rashid,

Jabeen

2024

Preprint

View full text Add to dashboard Cite

Background: Heart disease prediction is a critical healthcare task for identifying individuals at risk and enabling timely intervention. Accurate prediction models can help reduce morbidity and mortality rates associated with cardiovascular conditions. Various approaches, such as feature selection and hybrid models, have been proposed to improve the effectiveness and accuracy of the prediction of heart disease. Method: The study employs the Genetic Algorithm-Support Vector Machine-Convolutional Neural Network (GA-SVM-CNN) approach and evaluates it on three diverse datasets: UCI, Z-Alizadeh Sani, and Cardiovascular Disease Dataset. The genetic algorithm is utilized first to select the most relevant features from the datasets, effectively reducing dimensionality, eliminating irrelevant or redundant features, and choosing the most suitable ones. Subsequently, the hybrid SVM-CNN model is trained using the selected features, harnessing the complementary capabilities of both techniques to enhance prediction accuracy. Results: The performance of the GA-SVM-CNN approach is assessed using the three benchmark datasets and models. On the UCI dataset, the approach achieves an impressive accuracy of 98%, indicating its effectiveness in accurately predicting heart disease. On the Z-Alizadeh Sani dataset, the approach achieves an accuracy of 97%. On the Cardiovascular Disease Dataset, the approach achieves an accuracy of 86%. These high accuracy rates across different datasets underscore the efficacy of the GA-SVM-CNN approach in heart disease prediction. Conclusion: The combination of the genetic algorithm’s feature selection and the hybrid SVM-CNN model’s predictive power leads to superior performance in heart disease prediction. By accurately identifying individuals at risk of heart disease, this approach can enable timely interventions and contribute to more effective healthcare interventions.

show abstract

“…Reviews of JM methods have been described elsewhere [ 9 , 10 ]. However, papers using JM in CVD have been limited to a few longitudinal variables at most [ 11 ], and the capability of JM in handling higher-dimensional data remains unclear. Machine Learning (ML) methods have been the solution for higher-dimensional data.…”

Section: Introductionmentioning

confidence: 99%

Multivariate longitudinal data for survival analysis of cardiovascular event prediction in young adults: insights from a comparative explainable study

Nguyen

Vasconcellos

Keck

et al. 2023

BMC Med Res Methodol

View full text Add to dashboard Cite

Background Multivariate longitudinal data are under-utilized for survival analysis compared to cross-sectional data (CS - data collected once across cohort). Particularly in cardiovascular risk prediction, despite available methods of longitudinal data analysis, the value of longitudinal information has not been established in terms of improved predictive accuracy and clinical applicability. Methods We investigated the value of longitudinal data over and above the use of cross-sectional data via 6 distinct modeling strategies from statistics, machine learning, and deep learning that incorporate repeated measures for survival analysis of the time-to-cardiovascular event in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort. We then examined and compared the use of model-specific interpretability methods (Random Survival Forest Variable Importance) and model-agnostic methods (SHapley Additive exPlanation (SHAP) and Temporal Importance Model Explanation (TIME)) in cardiovascular risk prediction using the top-performing models. Results In a cohort of 3539 participants, longitudinal information from 35 variables that were repeatedly collected in 6 exam visits over 15 years improved subsequent long-term (17 years after) risk prediction by up to 8.3% in C-index compared to using baseline data (0.78 vs. 0.72), and up to approximately 4% compared to using the last observed CS data (0.75). Time-varying AUC was also higher in models using longitudinal data (0.86–0.87 at 5 years, 0.79–0.81 at 10 years) than using baseline or last observed CS data (0.80–0.86 at 5 years, 0.73–0.77 at 10 years). Comparative model interpretability analysis revealed the impact of longitudinal variables on model prediction on both the individual and global scales among different modeling strategies, as well as identifying the best time windows and best timing within that window for event prediction. The best strategy to incorporate longitudinal data for accuracy was time series massive feature extraction, and the easiest interpretable strategy was trajectory clustering. Conclusion Our analysis demonstrates the added value of longitudinal data in predictive accuracy and epidemiological utility in cardiovascular risk survival analysis in young adults via a unified, scalable framework that compares model performance and explainability. The framework can be extended to a larger number of variables and other longitudinal modeling methods. Trial registration ClinicalTrials.gov Identifier: NCT00005130, Registration Date: 26/05/2000.

show abstract

Modelling of longitudinal data to predict cardiovascular disease risk: a methodological review

Cited by 7 publications

References 117 publications

Tea Plant (Camellia sinensis): A Current Update on Use in Diabetes, Obesity, and Cardiovascular Disease

Tea Plant (Camellia sinensis): A Current Update on Use in Diabetes, Obesity, and Cardiovascular Disease

An Efficient Hybrid Classification Model for Heart Disease Prediction

Multivariate longitudinal data for survival analysis of cardiovascular event prediction in young adults: insights from a comparative explainable study

Contact Info

Product

Resources

About