Optimizing Bioimaging: Quantum Computing-Inspired Bald Eagle Search Optimization for Motor Imaging EEG Feature...

Area of Science:

• Neuroscience
• Bioimaging
• Computer Science

Background:

• Brain-computer interfaces (BCI) are crucial for neuroscientific research, particularly in analyzing electroencephalographic (EEG) signals.
• Effective feature selection is vital to reduce data dimensionality and improve BCI system performance by removing irrelevant or redundant information.

Purpose of the Study:

• To introduce a novel quantum computing-inspired bald eagle search optimization (QC-IBESO) method for enhancing motor imagery EEG feature selection.
• To improve classification accuracy and overcome the curse of dimensionality in BCI systems.

Main Methods:

• Utilized Z-score normalization for EEG data preprocessing.
• Applied Principal Component Analysis (PCA) for dimensionality reduction and feature extraction.
• Implemented the QC-IBESO algorithm for optimal EEG feature selection in motor imagery tasks.

Main Results:

• The QC-IBESO method effectively reduced EEG data dimensionality, facilitating the detection of critical motor imagery signals.
• The proposed approach demonstrated improved classification accuracy compared to conventional methods like neural networks, support vector machines, and logistic regression.
• Performance metrics including F1-score, precision, accuracy, and recall were computed to evaluate the method's efficacy.

Conclusions:

• The QC-IBESO method offers a novel and effective approach to EEG feature selection in bioimaging for BCI applications.
• This study highlights the potential of quantum-inspired optimization techniques in advancing neuroimaging and BCI research.