نوع مقاله : علمی-پژوهشی
نویسندگان
دانشگاه صنعتی سهند تبریز
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Brain-Computer Interfaces (BCIs) are transformative in neurorehabilitation, offering vital communication and control for individuals with severe motor impairments, such as those with ALS, spinal cord injuries, or stroke. By establishing direct links between brain activity and external devices, BCIs bypass damaged neural pathways, thereby restoring a degree of motor function and improving quality of life. Electroencephalography (EEG) is a leading modality for BCI development due to its accessibility and cost-effectiveness. However, a significant hurdle remains the inherent variability in cognitive and individual differences, which substantially impacts motor imagery (MI) task performance and BCI accuracy. This research introduces a novel approach for enhanced MI classification by specifically integrating the Common Spatial-Spectral Pattern (CSSP) filters with the Tunable-Q Wavelet Transform (TQWT). This synergistic combination, applied to the extensive CHO-2017 database (52 participants) which captures significant inter-individual cognitive variations, is designed to effectively address the challenges posed by individual differences in distinguishing between left and right-hand MI tasks. Critically, our method utilizes only the top 10 discriminative features extracted through this hybrid technique, significantly streamlining the process while maximizing classification efficacy. This tailored feature set demonstrated remarkable effectiveness across 99% of participants. When integrated with a K-Nearest Neighbors (KNN) classifier, this approach achieved an outstanding accuracy of 98.84%, surpassing current state-of-the-art methods. The findings of this research could pave the way for the development of more accurate BCI systems capable of extracting optimal commands for MI tasks.
کلیدواژهها [English]
)