Target position prediction system by machine learning corresponding to various irregular breathing patterns
| Project/Area Number |
17K10493
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Radiation science
|
|---|
| Research Institution | Keio University (2019)
Tokai University (2017-2018) |
|---|
Principal Investigator |
Kunieda Etsuo 慶應義塾大学, 医学部(信濃町), 客員教授 (70170008)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
株木 重人 東海大学, 医学部, 講師 (00402777)
藤田 幸男 東海大学, 医学部, 講師 (10515985)
松元 佳嗣 東海大学, 医学部, 助教 (20568969)
二上 菜津実 東海大学, 医学部, 助教 (20806195)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2019: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
|---|
| Keywords | 放射線治療 / 機械学習 / 胸部放射線治療 / 呼吸検出 / 標的位置予測 / 呼吸移動 / ファントム / 深層学習 / SBRT / 標的 / 位置予測 / 不規則呼吸 / 不規則呼吸パターン / Deep Neural Network / 動態ファントム |
|---|
| Outline of Final Research Achievements |
The purpose of this research is to develop a machine-learning system that analyzes the movement of the abdominal wall and predicts the target position for radiotherapy of the tumor such as a liver tumor. Deep Neural Network (DNN) was used to predict the relationship between abdominal wall movement and target position in various respiratory patterns. This allows the target position to be predicted even with irregular breathing. A newly developed dynamic motion phantom was used to evaluate our system. The phantom reproduces respiratory movement by increasing or decreasing the volume of air cells that simulates the lungs by means of air-compression drive. It is a completely new phantom that simulates the morphology of each organ. Furthermore, detailed analysis was performed to confirm the reproducibility of movements. In addition, several papers have been published on the outline of organs by machine learning of the chest and other radiation therapy related to respiration movements.
|
| Academic Significance and Societal Importance of the Research Achievements |
体幹部定位放射線治療は肝臓、膵臓などのX線で見えない部位への適応が広がっており、標的の呼吸移動に対する対応が重要となっている。現行の体表マーカーと4D-CTで知ることができる標的位置から両者の相関関係を求めて照射時の体表マーカーの動きを指標にして標的位置を予測する方法は、規則的、整った呼吸状態でないと予測が困難である。
腹壁全体の動きを３次元的、時間的に観察することによって腫瘍位置を予測できることに着目しディープラーニングを用いて、標的位置を予測することを発想した。本法を発展させ、普及すればより呼吸に影響される放射線治療の精度を高め治療成績を向上、また医療費の削減にも貢献できる。
|
Report
(4 results)
Research Products
(10 results)
-
-
-
-
-
-
-
-
[Presentation] Basic Study and Improvements of Electron-Tracking Compton Camera for Practical Use2017
Author(s)
S Kabuki, S Fujii , T Nagakura , J Yamashita , J Kushida , K Nishijima , A Takada , T MIzumoto , Y Mizumura9 , T Tanimori , E Kunieda
Organizer
American Association of Physicists in Medicine 59th annual meeting
Related Report
Int'l Joint Research
-
-
