Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Dec 26, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Stroke Prediction with Machine Learning Methods among Older Chinese.

Yafei Wu^1,2,3, Ya Fang^1,2,3

¹The State Key Laboratory of Molecular Vaccine and Molecular Diagnostics, School of Public Health, Xiamen University, Xiamen 361102, China.

International Journal of Environmental Research and Public Health

|March 18, 2020

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Differentiating cholangiocyte phenotype hepatocellular carcinoma and assessing early recurrence risk using contrast-enhanced ultrasound features and clinical indicators.

Abdominal radiology (New York)·2026

Same author

Multifunctional nano-platform for precision treatment of acute lung injury via synergistic anti-inflammation and oxidative stress modulation.

Journal of nanobiotechnology·2026

Same author

Dietary patterns and the progression from obesity to sarcopenic obesity in older adults: a prospective cohort study.

BMC geriatrics·2026

Same author

Global immunisation planning and economic development potential: a panel data analysis for the macroeconomic impact of hepatitis B vaccination from 2000 to 2018.

BMC medicine·2026

Same author

Life-course socioeconomic trajectories and cognitive aging in Chinese middle-aged and older adults.

Innovation in aging·2026

Same author

Gut microbial signatures for aging-related sarcopenia and dietary links among community-dwelling old-old adults: A metagenomic study.

Experimental gerontology·2026

Same journal

Correction: Grewal et al. Diversity and Representation in Cardiovascular Research: Evidence Gaps, Emerging Models, and Policy Implications. <i>Int. J. Environ. Res. Public Health</i> 2026, <i>23</i>, 241.

International journal of environmental research and public health·2026

Same journal

Drinking Water Quality and Health Risk Assessment in Rural Ghana: Evidence from North-East and North Gonja Districts in the Savannah Region.

International journal of environmental research and public health·2026

Same journal

Physical Activity of University Students During COVID-19 Restrictions: Evidence from Poland.

International journal of environmental research and public health·2026

Same journal

Assessment of Occupational Health and Safety Hazards in Mosquito Control Personnel in North Carolina and Virginia, USA.

International journal of environmental research and public health·2026

Same journal

Association Between Dysfunctional Parenting Practices and Suspected Gaming Disorder Among Japanese Male Junior High School Students: A Cross-Sectional Study of Parental Assessment.

International journal of environmental research and public health·2026

Same journal

A National Virtual Peer Support Group for Women Veterans Living with Breast Cancer: Lessons from the Field.

International journal of environmental research and public health·2026

See all related articles

Machine learning models effectively predict stroke in elderly Chinese populations using imbalanced data. Data balancing techniques significantly improved model performance, identifying key predictors like sex, hypertension, and uric acid.

Area of Science:

Computational Medicine
Epidemiology
Artificial Intelligence in Healthcare

Background:

Stroke is a prevalent condition requiring timely diagnosis and intervention.
Existing stroke prediction models often overlook the challenge of imbalanced datasets.
Imbalanced data, where stroke cases are rare compared to non-stroke cases, poses a significant hurdle for machine learning model development.

Purpose of the Study:

To develop and evaluate machine learning models for predicting stroke in an elderly Chinese population using imbalanced data.
To assess the effectiveness of data balancing techniques in improving stroke prediction model performance.
To identify key predictors of stroke within this demographic.

Main Methods:

Utilized data from a prospective cohort of 1131 elderly Chinese participants (2012-2014).

Keywords:

imbalanced data machine learning prediction stroke

More Related Videos

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

Published on: March 11, 2021

Related Experiment Videos

Last Updated: Dec 26, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

Digital Handwriting Analysis of Characters in Chinese Patients with Mild Cognitive Impairment

Published on: March 11, 2021

Applied data balancing techniques: random over-sampling (ROS), random under-sampling (RUS), and synthetic minority over-sampling technique (SMOTE).

Trained and compared regularized logistic regression (RLR), support vector machine (SVM), and random forest (RF) models.

Main Results:

Machine learning models performed poorly on imbalanced data, exhibiting very low sensitivity and AUC.
Data balancing techniques significantly improved model performance, with Random Forest achieving a maximum sensitivity of 0.78 and RLR an AUC of 0.72.
Sex, hypertension, and uric acid were identified as common predictors across all models; blood glucose, drinking, age, hs-CRP, and LDL-C were also significant.

Conclusions:

Machine learning models, when combined with data balancing techniques, are effective for stroke prediction in elderly populations with imbalanced data.
Data balancing is crucial for enhancing the sensitivity and accuracy of stroke prediction models in underrepresented groups.
The study highlights the potential of AI-driven approaches for improving stroke risk assessment and early intervention strategies.