| Project/Area Number |
13650463
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Measurement engineering
|
|---|
| Research Institution | Kyushu University (Faculty of Design) |
|---|
Principal Investigator |
FUKUSHIMA Shigehiro Kyushu University, Faculty of Design, Prof., 大学院・芸術工学研究院, 教授 (60027927)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YOSHINAGA Yukiyasu Kyushu University, Faculty of Design, Assistant Prof., 大学院・芸術工学研究院, 助手 (60304854)
|
|---|
| Project Period (FY) |
2001 – 2003
|
| Project Status |
Completed (Fiscal Year 2003)
|
| Budget Amount *help |
¥3,300,000 (Direct Cost: ¥3,300,000)
Fiscal Year 2003: ¥1,000,000 (Direct Cost: ¥1,000,000)
Fiscal Year 2002: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2001: ¥1,400,000 (Direct Cost: ¥1,400,000)
|
|---|
| Keywords | trend estimation / shading / hackground variation / compensating X-ray absorption / dynamical model / interpolation / computer-aided-diagnosis |
|---|
| Research Abstract |
This study attempted improving the quantification quality in medical X-ray imaging, an important application class, by compensating for the background variation, or trend, by estimating it from a projection image with transparently overlapping complex background. Our special but not limiting examples are double-contrast gastro-radiographic images. The expected effect is more definite correspondence between organ structures and tone-of-gray information, which may be useful in detection and quantification of legions as well as computer-aided and automatic diagnosis. This study specially focused on the physical principle of image formation. A fundamental concept has been composed in which the background is eatimated as a gray-level envelope, which interpolates in the image the gray levels of pixels with locally minimum absorption of the X-ray. This method makes it possible to discriminate the complex background structure in scales adaptive to organ and tissue structures under observation.
… More
The major fruits of this study include (1) that a method has been proposed in order to estimate additively overlapped background as a gray level envelope on the basis of a dynamic analogy of free fall of an elastic net, (2) that interpolation methods have been introduced by applying planar triangular patches and curved-surface patches in order to make the gray level envelope more globally adaptive, (3) that a frainework for estimating the background variation has been proposed as well as verifying the effect of the gray level compensation by using it, (4) that the pixel-sampling method has been proposed in order to improve the accuracy of the gray-level envelope, (5) that an advancement of the proposed method has been verified against a method based on the mathematical morphology, another method consistent with the transparently overlapping of X-ray images, (6) that an advancement of the proposed method has been verified against methods based on spatial frequency region filtering of a general approach in image signal processing, and (7) that the possibility has been experimentally demonstrated of estimating the three-dimensional shape structure on the basis of gray level information, as an example of quantification quality improvement of X-ray images. Less
|
