Scholarlink

<p>Stroke is an illness in which the brain is affected owing to lack of blood movement in arteries, producing injury to the brain. Due to lack of blood flow, other form of nutrients to the brain is broken up, in such case symptoms may be starting. From the information given by World Health Organization, the disease stroke is the highest reason of mortality worldwide. Primary identification of the several cautionary symptoms of a stroke can aid lessen the severity. Various machine learning (ML) models have been established to envisage the possibility of a stroke occurring in brain. Researchers use many machine learning algorithms, such as K-Nearest Neighbours, Support Vector Machine, Decision Tree, Random Forest, XGBoost (Extreme Gradient) and CatBoost to train data set for consistent calculation of stroke. Out of these ML techniques, Random Forest was the best with an accuracy of 99.2%. To develop this method an open-access Stroke Prediction dataset is used. The Percentage of accuracy of the proposed work is superior than the other existing works. The superiority of the proposed work is shown by numerous model comparisons.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Techniques for Stroke Prediction

  • 期刊论文

  • 发表日期：

    2026年2月13日

查看全文
  • B. Mathivanan,
  • P. Perumal

摘要

Stroke is an illness in which the brain is affected owing to lack of blood movement in arteries, producing injury to the brain. Due to lack of blood flow, other form of nutrients to the brain is broken up, in such case symptoms may be starting. From the information given by World Health Organization, the disease stroke is the highest reason of mortality worldwide. Primary identification of the several cautionary symptoms of a stroke can aid lessen the severity. Various machine learning (ML) models have been established to envisage the possibility of a stroke occurring in brain. Researchers use many machine learning algorithms, such as K-Nearest Neighbours, Support Vector Machine, Decision Tree, Random Forest, XGBoost (Extreme Gradient) and CatBoost to train data set for consistent calculation of stroke. Out of these ML techniques, Random Forest was the best with an accuracy of 99.2%. To develop this method an open-access Stroke Prediction dataset is used. The Percentage of accuracy of the proposed work is superior than the other existing works. The superiority of the proposed work is shown by numerous model comparisons.