Related Experiment Video
Updated: Jun 27, 2025
08:51
Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
1.2K
Exploring post-COVID-19 health effects and features with advanced machine learning techniques.
Muhammad Nazrul Islam1, Md Shofiqul Islam2, Nahid Hasan Shourav2
1Department of Computer Science and Engineering, Military Institute of Science and Technology, Mirpur Cantonment, Dhaka, 1216, Bangladesh. nazrul@cse.mist.ac.bd.
Scientific Reports
|April 30, 2024
Summary
Post-COVID-19 recovery is impacted by demographics and health factors like sleep and memory. Machine learning, particularly Decision Trees, effectively identifies key predictors of long-term health issues after COVID-19 infection.
Area of Science:
- Medical Research
- Public Health
- Data Science
Background:
- COVID-19, an infectious respiratory disease, presents diverse health outcomes, including long-term sequelae.
- Many recovered individuals experience persistent health issues influenced by demographic, physiological, and neurological factors.
Purpose of the Study:
- To investigate health factors affecting diverse demographic profiles in post-COVID-19 patients.
- To establish correlations among physiological and neurological factors in the post-COVID-19 state.
- To identify the most effective machine learning model for predicting post-COVID-19 health impacts.
Main Methods:
- Survey data collected from COVID-recovered patients in Bangladesh.
- Application of various machine learning algorithms to predict post-COVID-19 factors.
- Statistical validation using Chi-square and Pearson's coefficient.
- Feature importance analysis using Gini Index, Feature Coefficients, Information Gain, and SHAP Value Assessment.
Main Results:
- Significant relationships identified among post-COVID-19 health factors.
- Pearson's coefficient indicated associations among physiological and neurological factors.
- The Decision Tree model demonstrated superior performance in identifying crucial predictive features for post-COVID-19 impact.
Conclusions:
- Demographic and health factors significantly influence post-COVID-19 recovery and long-term health.
- Machine learning models, especially Decision Trees, are effective tools for understanding and predicting post-COVID-19 sequelae.
- Identifying key features aids in targeted interventions for improved patient outcomes.
