Machine Learning-Based Stroke Prediction

Weijun Deng, Chen Li, Zhirui Yan, Yuzhuo Yuan

2024

Abstract

The application of machine learning techniques in the medical diagnostic field has seen a gradual increase, with the search for an efficient and reasonably accurate prediction model becoming a focal point in related research areas. This study focuses on the comparison and evaluation of various machine learning models' performance on a stroke prediction dataset, aiming to identify the optimal prediction model. During the preliminary phase of the experiment, the dataset underwent preprocessing, which included handling missing values, label encoding of non-numeric data types, and feature selection based on the relevance between features and prediction labels. Moreover, models such as Logistic Regression, Decision Trees, XGBoost, and Random Forests were selected for in-depth analysis, and the Z-score method was employed for data normalization. Throughout the model tuning process, detailed model optimization was conducted through parameter adjustments and cross-validation methods. This study utilized AUC, precision0, and recall1 as evaluation metrics to conduct a comprehensive analysis of model performance, ultimately determining that the adjusted Random Forest and Logistic Regression models demonstrated the best performance in stroke prediction. The findings of this study provide an effective method for stroke prediction and offer guidance for future research in disease prediction using machine learning.

Paper Citation

in Harvard Style

Deng W., Li C., Yan Z. and Yuan Y. (2024). Machine Learning-Based Stroke Prediction. In Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI; ISBN 978-989-758-713-9, SciTePress, pages 754-761. DOI: 10.5220/0012972300004508

in Bibtex Style

@conference{emiti24,
author={Weijun Deng and Chen Li and Zhirui Yan and Yuzhuo Yuan},
title={Machine Learning-Based Stroke Prediction},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={754-761},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012972300004508},
isbn={978-989-758-713-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI
TI - Machine Learning-Based Stroke Prediction
SN - 978-989-758-713-9
AU - Deng W.
AU - Li C.
AU - Yan Z.
AU - Yuan Y.
PY - 2024
SP - 754
EP - 761
DO - 10.5220/0012972300004508
PB - SciTePress