期刊
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
卷 60, 期 1, 页码 90-96出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TBME.2012.2223698
关键词
Biomedical signal processing; EEG; medical diagnosis; traumatic brain injury (TBI)
资金
- National Science Foundation [CMMI-0845753]
- Department of Behavioral Science, University of Kentucky College of Medicine
- Div Of Civil, Mechanical, & Manufact Inn
- Directorate For Engineering [0845753] Funding Source: National Science Foundation
Traumatic brain injury (TBI) is the leading cause of death and disability in children and adolescents in the U. S. This is a pilot study, which explores the discrimination of chronic TBI from normal controls using scalp EEG during a memory task. Tsallis entropies are computed for responses during an old-new memory recognition task. A support vector machinemodel is constructed to discriminate between normal andmoderate/severe TBI individuals using Tsallis entropies as features. Numerical analyses of 30 records (15 normal and 15 TBI) show amaximum discrimination accuracy of 93% (p-value = 7.8557E-5) using four features. These results suggest the potential of scalp EEG as an efficacious method for noninvasive diagnosis of TBI.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据