1993 Fiscal Year Final Research Report Summary
Development of three-dimensionally measuring instrument of ocular fundus
| Project/Area Number |
04452105
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
物理計測・光学
|
|---|
| Research Institution | Kobe University |
|---|
Principal Investigator |
YOSHIMURA Takeaki Kobe University, Department of Instrumentation Engineering, Associate Professor, 工学部, 助教授 (70031127)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NAKAGAWA Kiyoshi Kobe University, Department of Instrumentation Engineering, Research Assosiate, 工学部, 助手 (50198032)
MINEMOTO Takumi Kobe University, Department of Instrumentation Engineering, Professor, 工学部, 教授 (60031093)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Keywords | 3-D Measurement of Ocular Fundus / Diagnosis of Glaucoma / Grating Projection Method / C / D Ratio and Rim Width / Visual Field Defects / Neural Network / ニューラルネットワーク |
|---|
| Research Abstract |
In order to achieve clinical diagnosis of glaucomatous eyes at early stage, new developments of the three dimensionally measuring ocular fundus camera and the data processing system of images obtained by a pattern projected method have been done in this reserch project. The results are shown as follows :
1. The improvements of the comercial ocular fundus camera (Kowa, type fx-50c) have achieved to take grating images of high contrast, even if its system is applied to any patients.
2. Developments of the image processing system have achieved to analyze automatically the contour map of an optic nerve head. Consequently, the standard deviation of the excavation depth was obtained within 15%.
3. By analysing two parameters of the cup-disc area ratio (C/D ratio) and the average depth of the cup area, it is found that the amount deduced from the two parameters does not depends on the different patients and can estimeate the grade of the glaucoma.
4. The narrowing of the rim width strongly depends on the field defects at the corresponding direction.
5. The neural network is possible to do the resulting estimation of 10 parameters, which is the rim widths at the 8 directions, the average depth and the C/D ratio. The results can estimate the glaucomatous eye at early stage without visual field defects.
|
