2016 Fiscal Year Annual Research Report
多値画像関数に基づく生物学的原体放射線治療最適計画システムの提案
| Project/Area Number |
16J04082
|
|---|
| Research Institution | Kyushu University |
|---|
Principal Investigator |
SOUFI MAZEN 九州大学, 医学系学府, 特別研究員(DC2)
|
|---|
| Project Period (FY) |
2016-04-22 – 2018-03-31
|
| Keywords | Radiation therapy / Differential geometry / Differential topology / Optimization / Range images |
|---|
| Outline of Annual Research Achievements |
An automated framework for detection of anatomical feature points on a patient’s surface in range images based on differential-geometry features for monitoring of patient positioning errors during the treatment time of high-precision radiation therapy has been developed. The range images were acquired for surfaces of two head phantoms, which were displaced by using a micrometer for simulation of patient positioning errors. Critical points, i.e. convex and concave anatomical feature points on two head phantoms were detected in range images acquired by a Kinect sensor with a submillimetre accuracy of 0.26±0.09 mm and 0.52±0.12 mm, respectively. Compared with previous studies, the developed framework could reduce the detection errors of the anatomical feature points.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
An automated system for detection of critical points, i.e. convex and concave feature points based on differential-geometry of a patient’s surface in range images has been developed for high-precision radiation therapy. Critical points have been detected on the head and neck phantom images with a submillimetre accuracy.
This research has been published in 2 international conferences (American Association of Physics in Medicine; The International Forum on Medical Imaging in Asia) and 1 domestic conference, and received the President Award of the Conference of Japan Society for Medical Physics (JSMP111, 2016). One original article has been published in an international journal (International Journal for Computer-Assisted Radiology and Surgery).
|
| Strategy for Future Research Activity |
Our purpose is to develop a differential-topology-based optimization planning system using multivariate image functions for high precision radiation therapy. However, current treatment strategies of cancer patients in radiation therapy are still not personalized and do not consider precision medicine criteria, which have been recognized essential for improving the treatment outcome. Therefore, we have decided to move toward the direction of development of a personalized radiation treatment planning using the differential topology and multivariate image functions. We are now investigating the development of an image-features-based treatment planning-support system in radiation therapy, which can predict the prognosis of lung cancer patients based on extensive image features.
|
Research Products
(5 results)
