2004 Fiscal Year Final Research Report Summary
Construction of awake background EEG model reflecting the EEG report
| Project/Area Number |
15560390
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Control engineering
|
|---|
| Research Institution | Fukuoka Institute of Technology |
|---|
Principal Investigator |
NISHIDA Shigeto Fukuoka Institute of Technology, Faculty of Information Engineering, Professor, 情報工学部, 教授 (10156078)
|
|---|
| Project Period (FY) |
2003 – 2004
|
| Keywords | background electroencephalogram / EEG report / EEG model / model parameter / automatic EEG interpretation / dominant rhythm / slow wave |
|---|
| Research Abstract |
Awake background electroencephalogram (EEG) is interpreted by electroencephalographer (EEGer), and the results of EEG interpretation are described in EEG reports. Desired EEG reports should include enough information so that the time series of the raw EEG can be imagined from them. Then, it is important to know how much information about the characteristics of EEG is included in the report.
In this study, a method for generating the time series reflecting the EEG report by use of an EEG model is proposed. First, "rhythmicity of slow waves", "period of amplitude variability of dominant rhythm" and "organization of dominant rhythm", those are important factors in the EEG interpretation, are represented quantitatively by use of EEG model parameters. By using these relationships between the model parameters and the characteristics of background EEG, an automatic EEG interpretation method by use of the EEG model is proposed. In this method, the EEG model taking account of trend artifacts is constructed to separate the trend artifacts from the background EEG. The proposed automatic EEG interpretation method was applied to actual EEG data, and gave the improved judgment. By using the automatic EEG interpretation method inversely, the parameters of the EEG model were appropriately determined based on the EEG report, and the model output was generated as the time series that reflects the EEG report. The proposed models were constructed from the EEG reports of 7 subjects. The constructed model time series was visually inspected by EEGer, and confirmed that they reflected the information in the EEG report for all subjects. By reproducing the time series from the EEG report, in which the summary of the characteristics of raw EEG is described, it is possible to know how much information is included in the EEG report.
|
