Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI

Abstract Classifying Electroencephalogram (EEG) signals for wheelchair navigation presents significant challenges due to high dimensionality, noise, outliers, and class imbalances. This study proposes an optimized classification framework that evaluates ten machine learning (ML) models, emphasizing...

Full description

Saved in:
Bibliographic Details
Main Authors: Amr M. Hamed, Abdel-Fattah Attia, Heba El-Behery
Format: Article
Language:English
Published: SpringerOpen 2025-07-01
Series:Journal of Big Data
Subjects:
Online Access:https://doi.org/10.1186/s40537-025-01238-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332968015790080
author Amr M. Hamed
Abdel-Fattah Attia
Heba El-Behery
author_facet Amr M. Hamed
Abdel-Fattah Attia
Heba El-Behery
author_sort Amr M. Hamed
collection DOAJ
description Abstract Classifying Electroencephalogram (EEG) signals for wheelchair navigation presents significant challenges due to high dimensionality, noise, outliers, and class imbalances. This study proposes an optimized classification framework that evaluates ten machine learning (ML) models, emphasizing ensemble methods, feature selection (FS), and outlier utilization. The dataset, comprising 2869 samples and 141 features, was processed using Recursive Feature Elimination (RFE) and correlation thresholds (CTs), achieving a peak accuracy of 69% with Extra Trees after FS. Notably, training on outlier-only data yielded even higher accuracy (Extra Trees: 82%), underscoring the value of outliers in enhancing class separability. Receiver Operating Characteristic–Precision Recall (ROC-PR) curve analysis confirmed that Extra Trees achieved a ROC AUC (Area Under Curve) of 0.92 and PR AUC of 0.82 for the best-classified movement command, while other models exhibited lower precision-recall (PR) balance. This approach, complemented by explainability techniques, offers a robust solution for EEG-based wheelchair control systems and paves the way for interpretable brain-computer interfaces (BCIs).
format Article
id doaj-art-3483fa17b4c24de0b88df81141b1e85c
institution Kabale University
issn 2196-1115
language English
publishDate 2025-07-01
publisher SpringerOpen
record_format Article
series Journal of Big Data
spelling doaj-art-3483fa17b4c24de0b88df81141b1e85c2025-08-20T03:46:03ZengSpringerOpenJournal of Big Data2196-11152025-07-0112113010.1186/s40537-025-01238-yOptimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AIAmr M. Hamed0Abdel-Fattah Attia1Heba El-Behery2Department of Computer Engineering and Systems, Faculty of Engineering, Kafrelsheikh UniversityDepartment of Computer Engineering and Systems, Faculty of Engineering, Kafrelsheikh UniversityDepartment of Computer Engineering and Systems, Faculty of Engineering, Kafrelsheikh UniversityAbstract Classifying Electroencephalogram (EEG) signals for wheelchair navigation presents significant challenges due to high dimensionality, noise, outliers, and class imbalances. This study proposes an optimized classification framework that evaluates ten machine learning (ML) models, emphasizing ensemble methods, feature selection (FS), and outlier utilization. The dataset, comprising 2869 samples and 141 features, was processed using Recursive Feature Elimination (RFE) and correlation thresholds (CTs), achieving a peak accuracy of 69% with Extra Trees after FS. Notably, training on outlier-only data yielded even higher accuracy (Extra Trees: 82%), underscoring the value of outliers in enhancing class separability. Receiver Operating Characteristic–Precision Recall (ROC-PR) curve analysis confirmed that Extra Trees achieved a ROC AUC (Area Under Curve) of 0.92 and PR AUC of 0.82 for the best-classified movement command, while other models exhibited lower precision-recall (PR) balance. This approach, complemented by explainability techniques, offers a robust solution for EEG-based wheelchair control systems and paves the way for interpretable brain-computer interfaces (BCIs).https://doi.org/10.1186/s40537-025-01238-yEEGBrain-computer interface (BCI)Wheelchair navigationMachine learningEnsemble methodsFeature selection
spellingShingle Amr M. Hamed
Abdel-Fattah Attia
Heba El-Behery
Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
Journal of Big Data
EEG
Brain-computer interface (BCI)
Wheelchair navigation
Machine learning
Ensemble methods
Feature selection
title Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
title_full Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
title_fullStr Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
title_full_unstemmed Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
title_short Optimization of EEG-based wheelchair control: machine learning, feature selection, outlier management, and explainable AI
title_sort optimization of eeg based wheelchair control machine learning feature selection outlier management and explainable ai
topic EEG
Brain-computer interface (BCI)
Wheelchair navigation
Machine learning
Ensemble methods
Feature selection
url https://doi.org/10.1186/s40537-025-01238-y
work_keys_str_mv AT amrmhamed optimizationofeegbasedwheelchaircontrolmachinelearningfeatureselectionoutliermanagementandexplainableai
AT abdelfattahattia optimizationofeegbasedwheelchaircontrolmachinelearningfeatureselectionoutliermanagementandexplainableai
AT hebaelbehery optimizationofeegbasedwheelchaircontrolmachinelearningfeatureselectionoutliermanagementandexplainableai