Using Machine Learning Algorithms of Stroke Prediction

Slavkovska, Daniela and Petreska, Anita and Ristevski, Blagoj and Nikolovski, Saso and Rendevski, Nikola (2024) Using Machine Learning Algorithms of Stroke Prediction. In: 14th International conference on Applied Internet and Information Technologies (AIIT2024), 8 November, 2024, Zrenjanin, Serbia.

This is the latest version of this item.

[thumbnail of AIIT2024 Proceedings-478-488.pdf] Text
AIIT2024 Proceedings-478-488.pdf - Published Version
Download (830kB)

Abstract

Stroke is a severe medical condition resulting from disrupted blood flow or ruptured blood vessels in the brain, often leading to life-threatening consequences. The World Health Organization (WHO) identifies stroke as a leading cause of death and disability worldwide.
Although significant research has focused on heart-related diseases, stroke prediction has received comparatively less attention. To address this gap, this paper presents machine learning models developed to predict stroke likelihood, utilizing key physiological factors associated with stroke risk. Six algorithms: logistic regression, decision tree, random forest, KNN, SVM and Naïve Baye, were implemented to train and test prediction models. The primary objective
was to determine the algorithm that provides the highest predictive accuracy.
Our findings reveal that the Naïve Bayes algorithm performed best, achieving an accuracy of approximately 82%. This is notable given Naïve Bayes’ suitability for probabilistic data and its efficiency in handling complex variable interactions, suggesting its value for early stroke
detection in clinical settings. The use of machine learning in stroke prediction highlights a promising approach for early intervention, potentially aiding in reducing stroke-related mortality and morbidity.
This paper contributes to expanding the application of machine learning in healthcare, emphasizing the need for focused stroke prediction research. Future work could enhance these models by integrating diverse datasets, testing additional machine learning techniques, and
refining predictive algorithms to boost accuracy and reliability. By advancing stroke prediction, machine learning may play a key role in mitigating stroke’s impact on global health.

Item Type: Conference or Workshop Item (Paper)
Subjects: Scientific Fields (Frascati) > Natural sciences > Computer and information sciences
Scientific Fields (Frascati) > Engineering and Technology > Electrical engineering, electronic engineering,information engineering
Divisions: Faculty of Information and Communication Technologies
Depositing User: Prof. d-r. Nikola Rendevski
Date Deposited: 16 Jan 2025 16:24
Last Modified: 16 Jan 2025 16:24
URI: https://eprints.uklo.edu.mk/id/eprint/10569

Available Versions of this Item

Actions (login required)

View Item View Item