| Abstrakt: |
Cerebral hemorrhage is a common disease of older adults, which could increase the risk of cognitive impairment. Electroencephalogram (EEG) characteristics can be analyzed to investigate the applied value in the assessment of cognitive impairment of the patients with cerebral hemorrhage. One hundred eighty-two patients (including patients with cognitive impairment [CHCI] and patients with cognitive normality [CHNC] with cerebral hemorrhage, and 120 normal healthy persons [control; CN]) were recruited between July 2008 to March 2012 at the department of neurology. All patients were analyzed by EEG, and analysis results were compared to the Montreal Cognitive Assessment (MoCA) scale, using the methods of correlation analysis, clustering analysis, and concordance analysis. The results indicated that patients with CHCI had significantly lower EEG beta power (0.814 ± 0.113 mcV2) relative to CHNC (1.601 ± 0.186 mcV2, P < .01) or CN group (1.713 ± 0.201 mcV2, P < .01). Significant negative correlation was found between the beta power and hemorrhage region, age, hemorrhage size, hemorrhage amount (r1= −.92223, r2= −.81084, r3= −.79258, r4= −.84961, respectively, all P < .001). There was good concordance between K-means clustering algorithm calculating the beta power and MoCA scoring (Kappa = 0.899, P < .001). In conclusion, the preliminary findings suggest that the recognition techniques of EEG hold considerable promise for the assessment of cognitive impairment post cerebral hemorrhage, which negatively related to the hemorrhage region, hemorrhage size, hemorrhage amount, and age. |
