Targeted Brain Computer Interface Utilisation by Employing Endogenous EEG Frameworks

Brain-computer interfaces (BCI) establish a direct communication link between the brain and computers or other external devices. These interfaces enhance human capabilities by either supplementing or replacing peripheral functions, with potential applications in fields like rehabilitation, affective computing, robotics, gaming, and neuroscience. Significant global research efforts have led to standardized platforms that address the challenges of complex, non-linear brain dynamics, feature extraction, and classification. However, time-varying psycho-neurophysiological fluctuations and their impact on brain signals present additional challenges in translating BCI technology from controlled laboratory settings to everyday use. This review provides an overview of recent advancements in the BCI field and outlines key challenges. In this paper, we propose a conceptual framework for personalized BCI applications, aimed at improving the user experience by tailoring services to individual needs and preferences based on endogenous electroencephalography (EEG) paradigms, including motor imagery (MI), speech imagery (SI), and visual imagery. The framework comprises two core components: user identification and intention classification, which allow for personalized services by identifying users and recognizing their intended actions through EEG signals. We validate the framework’s feasibility with a private EEG dataset from eight subjects, utilizing the ShallowConvNet architecture to decode EEG features. Experimental results show that user identification achieved an average classification accuracy of 0.996, while intention classification reached 0.55 accuracy across all paradigms, with MI showing the best performance. These results suggest that EEG signals can effectively support personalized BCI applications, offering strong user identification and reliable intention decoding, particularly for MI and SI.