The primary objective of this research is to enhance the classification accuracy of motor imagery (MI) electroencephalography signals to improve brain–computer interfaces (BCIs) for communication among individuals with mobility limitations. The main challenge is finding the correct frequency bands, efficient time-frequency representations, and accurately classifying these representations. Various classification models are available but have not achieved high accuracy, which is a key evaluation matrix. This research provides a method that first identifies the best frequency range utilizing several fast-converging optimization approaches. After filtering with the identified band, a continuous wavelet transform was used with complex Morlet to obtain temporal frequency representations, which are effective sources of feature representation for MI tasks. These scalograms were then classified using the advanced deep learning model vision transformer, which is well-known for its ability to extract and select features using the attention mechanism. The proposed technique achieved remarkable accuracy, attaining 97.33% on a widely recognized dataset and 89.89% on another dataset, outperforming comparable research. Integrating modern signal processing and a cutting-edge deep model enhances accuracy, allows for neuroprosthetic device control, and offers up new avenues for research in the BCI arena.

In the era of data, it is a challenging task to classify continuous data such as electroencephalographic data. The electroencephalographic signal maps several thoughts going in an individual's brain by connecting a device to the human brain. In this paper, we have proposed a deceit identification system using a test called “concealed information test.” The electroencephalographic data have been recorded when the concealed information test is performed for experimental analysis. To enhance the performance of the deceit identification system, the optimization of support vector machine (SVM) parameters and the selection of the EEG channels are performed. This paper implements a binary version of the BAT algorithm (binary BAT algorithm) and the conventional BAT algorithm on the electroencephalography (EEG) data. A novel cost function is also proposed, which utilizes the results of continuous BAT and binary BAT to enhance the system performance. In this synergistic approach, BAT is used for the SVM parameters optimization, and the binary BAT algorithm is applied for the EEG channel selection. The performance of the system is improved, and it is inferred that the channels placed at the occipital lobe of the brain consist of the artifacts. After removing the channels placed on the occipital lobe, i.e., O1, Oz, and O2, and using the optimized SVM parameters, the system's average accuracy increases from 94.11% to 96.8%.