Combining ICA Clustering and Power Spectral Density for Feature Extraction of Mental Fatigue of Spinal Cord Injury Patients

Publication Type:
Conference Proceeding
Citation:
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, 2019, pp. 530 - 533
Issue Date:
2019-07-01
Filename Description Size
Rifai_Chai_EMBC2019_Draft010.pdfAccepted manuscript840.77 kB
Adobe PDF
Full metadata record
© 2019 IEEE. This paper presents the combination of clustering-based independent component analysis (ICASSO) and power spectral density (PSD) as a features extractor of mental fatigue from spinal cord injury (SCI) patients. Initially, the results show that SCI and abled-bodied groups have no differences in EEG for alert and mental fatigue states. Further, the coefficient determination (R2) is calculated for testing the variation of data alert vs. fatigue on the SCI group, resulting in a lower R2 for proposed combination of ICASSO and PSD method compared to the PSD method only. With the lower R2 values, this shows that the proposed method ICASSO and PSD is able to provide superior distinction for separating fatigue vs. alert data variation. The statistical significance is found across four EEG bands and EEG channels.
Please use this identifier to cite or link to this item: