2023 Fiscal Year Final Research Report
Development of a novel cone-beam CT system for adaptive radiation therapy based on patient-specific treatment efficacy
| Project/Area Number |
19K17170
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 52040:Radiological sciences-related
|
|---|
| Research Institution | Yamaguchi University |
|---|
Principal Investigator |
Yuasa Yuki 山口大学, 医学部附属病院, 副診療放射・エックス線技師長 (20749840)
|
|---|
| Project Period (FY) |
2019-04-01 – 2024-03-31
|
| Keywords | Deep learning / DECT / CBCT / Iodine image |
|---|
| Outline of Final Research Achievements |
For the radiation therapy, cone-beam CT is used for patient set-up and observation of tumor and normal tissue. However, cone-beam CT has the issue of low soft tissue contrast and decreased image quality compared to diagnostic CT images. Therefore, we aimed to establish a dual-energy cone-beam CT system by integrating deep learning models with dual-energy technology. In this study, we developed a system that converts images obtained at a single tube voltage into CT images obtained at different tube voltages by inputting them into a deep learning model. This development enables the future expansion of dual-energy cone-beam CT system development.
|
| Free Research Field |
放射線科学
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究で開発したシステムは，医用画像処理と深層学習を応用することで，従来の手法では取得が難しかった単一管電圧で撮影されたCT画像から他管電圧で取得したCT画像の生成が可能となった。従来取得するCT画像を使用することで計算のみで追加画像を取得可能で、患者被ばく線量の低減に貢献することが可能である。また、本システムについては、既存の放射線機器に導入可能であり、画質の向上や画像診断制度の向上に寄与する可能性があると考えられる。
|
