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: Jan 7, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

HMLA: A hybrid machine learning approach for enhancing stroke prediction models with missing data imputation

M Sheetal Singh¹, Khelchandra Thongam¹, Krishna Kumar²

¹Computer Science and Engineering Department, NIT Manipur, Langol, Imphal, 795004, Manipur, India.

Scientific Reports

|December 20, 2025

Summary

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

From knowledge graph to topological data analysis: a novel framework to analyze gene regulatory networks for tomato-multi-pathogen interactions.

The New phytologist·2026

Same author

A Global Survey on Intra-operative Neuromonitoring (IONM) Practices, Attitudes Towards Its Use, and Barriers Among Spine Surgeons.

Global spine journal·2026

Same author

Amping Up Receptor Balance and Bias: Fine Tuning of Unimolecular Multiagonists.

Journal of medicinal chemistry·2026

Same author

Optimizing breast cancer treatment strategies through fractional-order dynamics: A computational modeling approach.

PloS one·2026

Same author

SedNet: A physics-informed operator-learning framework for rapid sedimentation velocity analytical ultracentrifugation analysis.

International journal of pharmaceutics·2026

Same author

Long-term trends and variability in sugarcane production: a five-district comparative analysis with meteorological context in Maharashtra and Karnataka, India.

Scientific reports·2026

This study introduces a hybrid machine learning model combining Information Gain Ratio (IGR), K-Nearest Neighbour (KNN), and Deep Neural Network (DNN) for accurate stroke risk prediction. The novel approach enhances data integrity and computational efficiency for early detection.

Area of Science:

Biomedical Informatics
Machine Learning in Healthcare
Cardiovascular Research

Background:

Early stroke prediction is vital for reducing mortality and disability.
Clinical datasets often contain irrelevant and sparse information, hindering predictive model performance.
Existing methods may struggle with data integrity and computational efficiency.

Purpose of the Study:

To develop and evaluate a novel machine learning framework for stroke risk prediction using the Cardiovascular Health Study (CHS) dataset.
To enhance data preprocessing by integrating feature selection and missing data imputation techniques.
To assess the predictive performance and computational efficiency of the proposed hybrid model.

Main Methods:

Feature selection using Information Gain Ratio (IGR) for preprocessing.

Keywords:

Deep neural network Feature selection Missing data imputation Stroke prediction

More Related Videos

Compensatory Limb Use and Behavioral Assessment of Motor Skill Learning Following Sensorimotor Cortex Injury in a Mouse Model of Ischemic Stroke

Compensatory Limb Use and Behavioral Assessment of Motor Skill Learning Following Sensorimotor Cortex Injury in a Mouse Model of Ischemic Stroke

Published on: July 10, 2014

Related Experiment Videos

Last Updated: Jan 7, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Compensatory Limb Use and Behavioral Assessment of Motor Skill Learning Following Sensorimotor Cortex Injury in a Mouse Model of Ischemic Stroke

Compensatory Limb Use and Behavioral Assessment of Motor Skill Learning Following Sensorimotor Cortex Injury in a Mouse Model of Ischemic Stroke

Published on: July 10, 2014

Missing data imputation and data integrity enhancement using K-Nearest Neighbour (KNN).

Stroke risk classification using a Deep Neural Network (DNN) model.

Model validation through a 10-fold nested cross-validation scheme.

Main Results:

The hybrid IGR-KNN-DNN framework achieved high performance metrics: 94.32% accuracy, 95.96% precision, 95.00% F1-score, 94.67% specificity, and 94.06% sensitivity.
The approach demonstrated strong predictive potential and computational efficiency.
Comparison with classical methods confirmed the model's effectiveness.

Conclusions:

The hybrid IGR-KNN-DNN framework shows significant promise for early stroke-risk assessment.
The model's preprocessing steps enhance data integrity and computational efficiency.
Further external validation is recommended to broaden clinical applicability.