A comparative analysis of machine learning classifiers for stroke prediction: A predictive analytics approach
Problem
Accurate early prediction of stroke is needed to enable timely treatment, but medical data is imbalanced.
Why it matters
Stroke is the third leading cause of death worldwide; early prediction could reduce mortality.
What the AI did
Eleven machine learning classifiers were trained and compared on stroke data using random oversampling, hyperparameter tuning, and cross-validation.
Results
Ten classifiers exceeded 90% accuracy before balancing; four exceeded 96% after oversampling.
Impact
Demonstrates high accuracy on study data; clinical deployment not stated in source.
Record
| Venue | Healthcare Analytics |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Nitish Biswas, Khandaker Mohammad Mohi Uddin, Sarreha Tasmin Rikta, Samrat Kumar Dey |
| Citations | 116 (OpenAlex, at import time) |
| References | 50 |
| Model or method | SVM, Random Forest, KNN, Decision Tree, Naive Bayes, Voting Classifier, AdaBoost, Gradient Boosting, MLP, Nearest Centro |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1016/j.health.2022.100116 |
| Open access | gold |
| Keywords | stroke prediction, machine learning, classifiers, oversampling, imbalanced data |
Importance rating
existential / global scale
Abstract states stroke is the third leading cause of death in the world, causing severe illness, disability, or death; early accurate prediction could reduce mortality, matching leading-global-cause-of-death scale.