Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Comparative Analysis of Stroke Prediction Models Using Machine Learning

Anastasija Tashkova, Stefan Eftimov, Bojan Ristov, Slobodan Kalajdziski

TL;DR

This study tackles stroke risk prediction using machine learning on a dataset with significant class imbalance. By evaluating five algorithms (LR, RF, DT, SVM, XGBoost) and applying oversampling, SMOTE, and undersampling, it demonstrates that ensemble methods and SVM achieve the best predictive performance, though sensitivity for positive stroke cases remains a challenge. It also conducts a detailed feature-importance analysis, finding that Age, Average Glucose, and BMI are the strongest predictors, with SHAP confirming their impact and revealing age-related shifts in predictor importance for the elderly. The work highlights practical implications for clinical decision support and emphasizes the need for balanced, interpretable models to enable reliable early intervention; it also shows that age-specific models can improve relevance for older patients. Overall, the paper advances stroke risk assessment tools by combining rigorous data preprocessing, comprehensive model comparison, and nuanced feature interpretation.

Abstract

Stroke remains one of the most critical global health challenges, ranking as the second leading cause of death and the third leading cause of disability worldwide. This study explores the effectiveness of machine learning algorithms in predicting stroke risk using demographic, clinical, and lifestyle data from the Stroke Prediction Dataset. By addressing key methodological challenges such as class imbalance and missing data, we evaluated the performance of multiple models, including Logistic Regression, Random Forest, and XGBoost. Our results demonstrate that while these models achieve high accuracy, sensitivity remains a limiting factor for real-world clinical applications. In addition, we identify the most influential predictive features and propose strategies to improve machine learning-based stroke prediction. These findings contribute to the development of more reliable and interpretable models for the early assessment of stroke risk.

Comparative Analysis of Stroke Prediction Models Using Machine Learning

TL;DR

This study tackles stroke risk prediction using machine learning on a dataset with significant class imbalance. By evaluating five algorithms (LR, RF, DT, SVM, XGBoost) and applying oversampling, SMOTE, and undersampling, it demonstrates that ensemble methods and SVM achieve the best predictive performance, though sensitivity for positive stroke cases remains a challenge. It also conducts a detailed feature-importance analysis, finding that Age, Average Glucose, and BMI are the strongest predictors, with SHAP confirming their impact and revealing age-related shifts in predictor importance for the elderly. The work highlights practical implications for clinical decision support and emphasizes the need for balanced, interpretable models to enable reliable early intervention; it also shows that age-specific models can improve relevance for older patients. Overall, the paper advances stroke risk assessment tools by combining rigorous data preprocessing, comprehensive model comparison, and nuanced feature interpretation.

Abstract

Stroke remains one of the most critical global health challenges, ranking as the second leading cause of death and the third leading cause of disability worldwide. This study explores the effectiveness of machine learning algorithms in predicting stroke risk using demographic, clinical, and lifestyle data from the Stroke Prediction Dataset. By addressing key methodological challenges such as class imbalance and missing data, we evaluated the performance of multiple models, including Logistic Regression, Random Forest, and XGBoost. Our results demonstrate that while these models achieve high accuracy, sensitivity remains a limiting factor for real-world clinical applications. In addition, we identify the most influential predictive features and propose strategies to improve machine learning-based stroke prediction. These findings contribute to the development of more reliable and interpretable models for the early assessment of stroke risk.
Paper Structure (22 sections, 4 figures, 1 table)

This paper contains 22 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Global Burden of Stroke
  3. Machine Learning in Healthcare
  4. Research Gap and Objectives
  5. Data Preprocessing Pipeline
  6. Dataset Description
  7. Feature Categories
  8. Handling Missing Values
  9. Categorical Variable Encoding
  10. Addressing Class Imbalance
  11. Analysis of Key Variables
  12. Machine Learning Models and Optimization
  13. Model Selection
  14. Hyperparameter Tuning
  15. Results and Discussion
  16. ...and 7 more sections

Figures (4)

  • Figure 1: Exploratory Data Analysis (EDA) of Stroke Prediction Dataset
  • Figure 2: Summary results from models
  • Figure 3: XGBoost Model Feature Importance
  • Figure 4: SHAP summary plots for the XGBoost model