| Project/Area Number |
06557027
|
|---|
| Research Category |
Grant-in-Aid for Developmental Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Hygiene
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
KOYAMA Akio Department of Hygiene and Preventive Medicine, Hokkaido University School of Medicine, Associate professor., 医学部, 助教授 (90215202)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
EGUCHI Eiji Neuro Science Co. Ltd. President., 代表取締役
TAKAKURA Masayuki Department of Hygiene and Preventive Medicine, Hokkaido University School of Med, 医学部, 助手 (90241314)
SAITO Takeshi Department of Hygiene and Preventive Medicine, Hokkaido University School of Med, 医学部, 講師 (40153811)
SAITO Kazuo Department of Hygiene and Preventive Medicine, Hokkaido University School of Med, 医学部, 教授 (80000917)
|
|---|
| Project Period (FY) |
1994 – 1995
|
| Project Status |
Completed (Fiscal Year 1995)
|
| Budget Amount *help |
¥12,100,000 (Direct Cost: ¥12,100,000)
Fiscal Year 1995: ¥2,000,000 (Direct Cost: ¥2,000,000)
Fiscal Year 1994: ¥10,100,000 (Direct Cost: ¥10,100,000)
|
|---|
| Keywords | Image analysis / Visualization / Electrical anctivity in brain / Time series analysis / Neural network / 時系列解除 |
|---|
| Research Abstract |
This study intended to develop a newly devised software sysytem for three dimensional imaging analysis of electrical activity in the brain.
The role of the software system was to represent two dimensinal topographical mappings from 12 channels EEG recording data using maximum entropy spectral analysis (MEM), and the three-dimensional (3D) graphics compressed of these many topographical mappings accompanying time process. The software system was mainly devised from the combination of two application programs on the market, MemCalc system (for MEM and chaos analysis) and Voxel View system (for 3D imaging analysis), and an original program for compiling the data format.
The first step results confirmed that MEM analysis had a superior performance for detecting sufficient power spectrum from biological time series data without any kinds of modification to the native data, and that successful topographical mappings with higher resolution and stability were obtained using MEM than using the popular FFT analysis. Further, as a result of the second step, a new software system providing clear 3D images compressed many mapping data sets drawn as X-axis and Y-axis adding time process as Z-axis was developed. To evaluate the performance of resolution of this system, we demonstrated 3D imaging analysis of time series graphical data of electrical activities in a rat brain using other programs on the market.
These results suggest that our new system is useful to recognize clearly and instantly localized changes of electrical activities in the brain during the time process by only a few graphics, and has the possibility for application of nonlinear analysis for clarifying the mechanisms of time series changes.
|
